Skip to main content
Devin Desktop (formerly Windsurf) SignalsBack to the read

Every source behind the read

Every source GILD analyzed, newest first. Each says who the source is, how it knows, whether it still applies and which findings it counts toward. Open a count to see only its sources.

Whose sources count

Source

106 sources in All practitioners, with product documentation Showing all 106 sources

  1. Negative

    princevegeta89, Hacker News commenter

    Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraHacker News comment

    A long-time Windsurf user, the commenter says the product has fallen several features behind Cursor since becoming Devin, and has gone back to VS Code with Copilot Pro.

    Published Sep 11, 2026self-described: Long-time Windsurf user

    Practitioner account, pseudonymous; direct use; counts toward Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-09-11
    Counts toward
    Adoption and upkeep, continuity: unfavorable, G1 (1 unit before bounds), capped: C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current

  2. Mixed

    fishtoaster, Hacker News commenter

    Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraHacker News comment

    Enjoying Devin's cloud agents after comparing alternatives, the commenter calls the gap between Devin cloud and Devin Desktop embarrassing, describing Desktop as a thinly reskinned Windsurf.

    Published Sep 11, 2026self-described: Solo; evaluated cloud agents after leaving Cursor

    Practitioner account, pseudonymous; opinion; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Opinion
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Opinion, no use stated
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-09-11
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current

  3. Mixed

    bluelightning2k, Hacker News commenter

    Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraHacker News comment

    Still uses Devin Desktop, formerly Windsurf, as their editor but has not used its agent for a while because Codex now handles that work for them.

    Published Sep 10, 2026self-described: Uses Devin Desktop as their editor

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-09-10
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current

  4. Mixed

    PolCPP, Hacker News commenter

    Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraHacker News comment

    Grandfathered on old pricing, the commenter used Windsurf as their main editor until the pricing model changed and now uses it only to spend their weekly tokens on frontier models.

    Published Sep 10, 2026self-described: Grandfathered individual subscriber

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Observed on fable/astra
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-09-10, fable/astra
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current

  5. Mixed

    hightrix, Hacker News commenter

    Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraHacker News comment

    Their company has run it on an enterprise plan for about nine months. The commenter rates it above Copilot but now finds Claude and Codex better, and Devin's web and desktop harnesses inconsistent.

    Published Sep 10, 2026self-described: Their company, on an enterprise plan for about nine months

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Observed on enterprise plan
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-09-10, enterprise plan
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Current
    Observed on
    Enterprise plan

  6. Negative

    paimapi, Hacker News commenter

    Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraHacker News comment

    When the product was still Windsurf, terminal runs hung or froze nearly every time until the commenter wired up a workaround skill; the commenter says its multi-agent UI lead has since disappeared.

    Published Sep 10, 2026self-described: Former Windsurf user

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-09-10
    Not counted
    Reliability, run stability: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  7. Mixed

    RussianCow, Hacker News commenter

    Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceHacker News comment

    After under a month on Devin the commenter hits the US$20 plan's limits faster than with Claude Code or Codex, and rates Cursor's subscription the better value.

    Published Jul 9, 2026self-described: Solo, under a month on Devin

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Observed on $20/month plan
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-07-09, $20/month plan
    Not counted
    Cost predictability and limits, cost predictability: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current
    Observed on
    US$20 Devin plan

  8. Positive

    fallinditch, Hacker News commenter

    Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceHacker News comment

    Uses Cognition's SWE-1.6 model for grunt work alongside Opus for planning and harder problems, and is pleased with Cognition's progress since it took over Windsurf.

    Published Jul 8, 2026self-described: Former Windsurf user, now on Devin Desktop

    Practitioner account, pseudonymous; direct use; counts toward Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Observed on SWE 1.6
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-07-08, SWE 1.6
    Counts toward
    Adoption and upkeep, continuity: favorable, G1 (1 unit before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current
    Observed on
    SWE-1.6 and Opus

  9. Negative

    godzillabrennus, Hacker News commenter

    Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceHacker News comment

    Their organization was a Windsurf customer at the acquisition, then left after Cognition stopped offering support, raised prices twice and dismantled the brand.

    Published Jul 8, 2026self-described: Their organization, a former Windsurf customer

    Practitioner account, pseudonymous; direct use; counts toward Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-07-08
    Counts toward
    Adoption and upkeep, continuity: unfavorable, G1 (1 unit before bounds), capped: C-relationship
    Not counted
    Cost predictability and limits, cost predictability: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Current

  10. Positive

    RussianCow, Hacker News commenter

    Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceHacker News comment

    Runs a spec-driven workflow in Devin Desktop over ACP: Claude Code for exploration, Devin's SWE model for building and GLM 5.2 for verification, with little extra effort.

    Published Jul 8, 2026self-described: Solo, spec-driven workflow

    Practitioner account, pseudonymous; direct use; counts toward Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Observed on Devin SWE
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-07-08, Devin SWE
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current
    Observed on
    Devin Desktop with ACP agents

  11. Mixed

    settled, Hacker News commenter

    Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceHacker News comment

    Found SWE-1.5 and 1.6 fine for grunt work and tests, but says paid-model usage has been cut sharply since the US$15 plan, and a Codex subscription is now better value.

    Published Jul 8, 2026self-described: Solo; used Windsurf before the rename

    Practitioner account, pseudonymous; direct use; counts toward Cost predictability and limits How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Observed on SWE-1.5 and 1.6; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-07-08, SWE-1.5 and 1.6
    Counts toward
    Cost predictability and limits, cost transparency: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group
    Not counted
    Cost predictability and limits, limits in practice: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current
    Observed on
    SWE-1.5 and SWE-1.6

  12. No opinion stated

    Jeff Davis, VP of Core Infrastructure at Indeed

    2x the power users: How structured AI training scaled developer productivityConference session page with transcript, DX Annual via Engineering Enablement by DX

    Davis reports that around the start of 2025 Indeed had Cody, Copilot, unlimited budgets and training, yet only about 25% of developers used AI tools weekly. Co-presenter Michael Redding reports that weekly use reached around 97%, mostly Claude Code, a little Cursor, some Windsurf and Amp, with developers using several tools at once.

    "…our breakdown rate, we’re mostly Claude Code, a little bit Cursor, some Windsurf, and then down at the bottom, Amp."
    Excerpt, Jeff Davis

    Published Jun 29, 2026Job site with more than 2,000 engineers; talk given with product manager Michael Redding

    Published on
    Podcasts, Talks & Videos
    Who
    Technical leadership
    Evidence
    Measured
    Applies to
    Current
    Observed on
    Weekly AI tool usage across Indeed engineering

  13. Positive

    Jason Valentino, Head of Software Engineering Strategy at BNY

    Mapping the new SDLC at BNY: Codifying AI into every step of the delivery lifecycleConference session page with transcript, DX Annual via Engineering Enablement by DX

    Valentino reports that BNY engineers have Windsurf, Copilot and Claude Code. Today peer review means using the peer review agent in Windsurf or Claude Code, which "makes life better" but is still human driven; the goal is a deterministic evaluation when a merge request hits.

    "Right now, peer review is something that like, “Hey, you got Windsurf or you have ClaudeCode installed, you used the peer review agent, it makes life better.” It’s still human driven."
    Excerpt, Jason Valentino

    Published Jun 8, 2026Bank with more than 8,000 engineers; owns developer tooling and DevEx

    Counts toward

    Published on
    Podcasts, Talks & Videos
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Current
    Observed on
    Windsurf, Copilot and Claude Code installed across BNY engineering

  14. Positive

    Maksim Danilchenko, Engineering Lead at inDrive

    Devin Desktop (ex-Windsurf): Features, Pricing, What ChangedPersonal blog post, danilchenko.dev

    Danilchenko measured Devin Local at roughly 18,000 tokens against about 24,000 for Cascade on a 14-file refactor and saw subagents split work "maybe 40% of the time on multi-step tasks". Calls Devin Local measurably better than Cascade, but says the acquisition story creates real trust risk.

    Published Jun 6, 2026self-described: Hands-on review after the rename; personal blog

    Practitioner account, self-described; direct use; counts toward Delivered outcomes, Reliability, Operational control, Cost predictability and limits, Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Artifact checked (benchmark output); Observed on Devin Local
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    1 artifact (benchmark output)
    Date and version
    2026-06-06, Devin Local
    Counts toward
    Delivered outcomes, measured outcomes: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Reliability, run stability: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Cost predictability and limits, cost transparency: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Adoption and upkeep, setup: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Adoption and upkeep, continuity: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group
    Not counted
    Cost predictability and limits, limits in practice: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Measured
    Applies to
    Current
    Observed on
    Devin Desktop, Devin Local vs Cascade, SWE-1.6

  15. Positive

    Edwar Diaz, developer

    From Windsurf to Devin Desktop: my first impressions after migratingPersonal blog post, DEV Community

    After more than a year on Windsurf, their move to Devin Desktop kept their extensions and settings, and Copilot CLI running over ACP picked up and resumed their existing sessions.

    Published Jun 4, 2026self-described: Solo; more than a year on Windsurf

    Practitioner account, self-described; direct use; counts toward Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Specific details
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-06-04
    Counts toward
    Adoption and upkeep, setup: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Adoption and upkeep, continuity: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current
    Observed on
    Devin Desktop with Copilot CLI over ACP

  16. Mixed

    Ravi Patel, solo founder

    Windsurf Review 2026: Not For Solo Founders, Great For Small TeamsPersonal blog post, DEV Community (from rikuq.com)

    Tried Windsurf's free tier for about a week when the Claude weekly quota ran out, and found tab completion fast and Cascade reliably executing small-to-medium refactors. Calls it good but didn't keep it: its target user isn't a solo agentic-AI shipper, and the paid tier adds Opus access already had directly.

    Published May 25, 2026self-described: Solo founder; one-week trial on the free tier

    Practitioner account, self-described; direct use; counts toward Delivered outcomes, Reliability, Operational control, Cost predictability and limits How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Observed on free tier; Specific details
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-05-25, free tier
    Counts toward
    Delivered outcomes, task completion: favorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, context handling: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Reliability, run stability: favorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Operational control, models and deployment: favorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Cost predictability and limits, cost transparency: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Cost predictability and limits, limits in practice: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Free tier

  17. Negative

    isqueiros, Hacker News commenter at a company with company-wide Windsurf accounts

    Comment on: Changes to GitHub Copilot individual plansHacker News comment

    Their company holds company-wide Windsurf accounts, yet the commenter says nearly everyone uses Claude Code through Bedrock instead.

    Published Apr 22, 2026self-described: Company with company-wide Windsurf, Copilot and ChatGPT Enterprise accounts

    Practitioner account, pseudonymous; second-hand; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Second-hand
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Reports others' experience
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-04-22
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Organization-wide
    Evidence
    Reported
    Applies to
    Still applicable

  18. Mixed

    zazibar, Hacker News commenter at a company with over 400 engineers

    Comment on: Anthropic downgraded cache TTL on March 6thHacker News comment

    Their 400-engineer company canceled all IDE subscriptions, Windsurf included, and moved everyone to Claude Code as a cost-saving measure with no migration plan. Token spend then forced limits.

    Published Apr 12, 2026self-described: Company of over 400 engineers that canceled its Windsurf subscriptions

    Practitioner account, pseudonymous; second-hand; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Second-hand
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Reports others' experience
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-04-12
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Organization-wide
    Evidence
    Reported
    Applies to
    Still applicable

  19. Negative

    visarga, Hacker News commenter

    Comment on: Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClawHacker News comment

    Left Windsurf, and later Cursor, after each cut their usage quota within weeks or months of their paying for a plan; the commenter never returned to Windsurf.

    Published Apr 4, 2026self-described: Former Windsurf subscriber

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-04-04
    Not counted
    Cost predictability and limits, cost predictability: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  20. Positive

    Patt_, Hacker News commenter

    Comment on: Ask HN: US$50 monthly budget, which coding models would you recommend now?Hacker News comment

    On a US$50 monthly budget the commenter pays US$20 for Windsurf for access to nearly all models, reserving Claude Opus and Sonnet for tricky work once requirements are clear.

    Published Mar 24, 2026self-described: Solo on a US$50 monthly tooling budget

    Practitioner account, pseudonymous; direct use; counts toward Cost predictability and limits How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Observed on Claude Opus/Sonnet; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-03-24, Claude Opus/Sonnet
    Counts toward
    Cost predictability and limits, cost transparency: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Operational control, models and deployment: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    US$20 Windsurf plan

  21. Positive

    dennisjoseph, Hacker News commenter

    Comment on: Ask HN: Who is still using Windsurf and why?Hacker News comment

    Keeps Windsurf as their primary IDE because bring-your-own-key covers prompts, agents and autocomplete, including in Vim, which the commenter says Cursor and Zed do not offer.

    Published Mar 20, 2026self-described: Solo; Windsurf is their primary IDE

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-03-20
    Not counted
    Operational control, models and deployment: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  22. Mixed

    oib, Windsurf user filing a bug report

    Windsurf Bug Report - Memory Leak Consuming All RAM on LinuxGitHub issue

    Reports that Windsurf on Debian consumed all 64 GB of RAM after startup until it crashed, with no workaround found.

    Published Mar 10, 2026self-described: Solo, Debian 13 with GNOME

    Practitioner account, pseudonymous; direct use; counts toward Reliability How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Observed on 1.9577.24-1773087928
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-03-10, 1.9577.24-1773087928
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf 1.9577.24 on Debian 13

  23. Negative

    sensanaty, Hacker News commenter

    Comment on: Labor market impacts of AI: A new measure and early evidenceHacker News comment

    Tried Windsurf's Arena Mode on a test-completion task, with AGENTS.md and rules files in place, and saw the same failure as in Claude Code; the commenter reviews every command it runs.

    Published Mar 6, 2026self-described: Engineer at a company with AGENTS.md and rules files in place

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-03-06
    Counts toward
    Delivered outcomes, task completion: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: unfavorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  24. Mixed

    Saloni Agrawal, React and React Native developer

    I Tried Windsurf: An AI Native Coding EnvironmentPersonal blog post, DEV Community

    A long-time VS Code user, Agrawal liked queuing several instructions while Cascade works and its persistent memory, but finds the overlap with VS Code agent mode significant.

    Published Mar 4, 2026self-described: Solo; VS Code user trying Windsurf

    Practitioner account, self-described; direct use; counts toward Delivered outcomes, Reliability, Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Specific details
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2026-03-04
    Counts toward
    Delivered outcomes, correctness: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Delivered outcomes, context handling: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Reliability, run stability: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Operational control, execution control: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  25. Mixed

    aeongeon, Windsurf user filing a bug report

    # Windsurf IDE Freeze Bug ReportGitHub issue

    Measured the editor freezing for about five minutes after every Cascade response on Linux, traced with strace to a blocked renderer thread and dozens of duplicate calls.

    Published Feb 10, 2026self-described: Solo, Linux (Pop!_OS)

    Practitioner account, pseudonymous; structured test; counts toward Reliability, Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Ran a structured test; Artifact checked (log); Observed on 1.9544.35 (1.106.0)
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Ran a structured test
    Support
    1 artifact (log)
    Date and version
    2026-02-10, 1.9544.35 (1.106.0)
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Operational control, data control: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Measured
    Applies to
    Needs a recheck
    Observed on
    Windsurf 1.9544.35 on Linux

  26. No opinion stated

    James Reggio, chief technology officer, Brex

    Latent Space: Brex's AI Hail Mary, with James ReggioPodcast episode with publisher transcript

    Brex lets employees pick between Cursor, Windsurf and Claude Code credits and switch easily, buying a few seats of each.

    "You can pick, you know, between like Cursor, Windsurf, Cloud Code, Credits, like, and you can basically craft your stack to your preference and easily switch between them."
    Transcript excerpt, James Reggio, around 21:15 in the episode

    Published Jan 17, 2026self-described: Brex, a few seats of several tools chosen by employees

    Published on
    Podcasts, Talks & Videos
    Who
    Organization-wide
    Evidence
    Reported
    Applies to
    Current

  27. No opinion stated

    Sina Pakazad, Vice President, Data Science, C3 AI

    C3 AI: Autonomous coding agents beyond developer productivityOrganization write-up

    C3 AI surveyed coding agents including Windsurf, Cursor, Devin, Codex and Claude Code, and chose OpenHands Cloud for its remote programmability and scale.

    "Among existing ACAs, OpenHands (Cloud) stands out for providing full remote programmability and unconstrained interaction in the cloud, resulting in unlimited scalability."
    Excerpt, Sina Pakazad

    Published Dec 24, 2025self-described: C3 AI, coding agents inside its platform

    Counts toward

    Published on
    Engineering Blogs & Reviews
    Who
    Organization-wide
    Evidence
    Reported
    Applies to
    Current

  28. Positive

    Darryn Dieken, Chief Availability Officer at Salesforce

    How AI-Enabled Tooling Boosted Code Output 30% While Keeping Quality and Deployment Safety IntactCompany engineering blog post, Salesforce Engineering

    Salesforce offers Cursor and Windsurf for code and test generation, with over 85% of eligible engineers using AI tools weekly; code in production rose 30%, which pushed the bottleneck to review and testing.

    "A few tools in this portfolio include: Cursor and Windsurf for AI-assisted code and tests generation…"
    Excerpt, Darryn Dieken

    Published Dec 16, 2025Engineering productivity and reliability for thousands of Salesforce engineers across 25 Hyperforce regions

    Practitioner account, name and role on their organization's own page; ran the rollout; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Name and role on their organization's own page; Customer of Devin Desktop; Ran the rollout
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (Darryn Dieken, Chief Availability Officer)
    Relationship
    customer (self disclosed); customer (self disclosed); customer (self disclosed)
    How they know
    Ran the rollout
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-12-16
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Engineering Blogs & Reviews
    Who
    Technical leadership
    Evidence
    Measured
    Applies to
    Still applicable
    Observed on
    Windsurf and Cursor in a portfolio of AI coding tools

  29. Positive

    gcbirzan, Hacker News commenter whose team moved from Copilot to Windsurf

    Comment on: Ask HN: How can I get better at using AI for programming?Hacker News comment

    Their team moved to Windsurf from Copilot, and the author finds it most useful for answering questions about a monorepo of a bit over 1M LoC; the supertab feature was also very useful. Almost never asks it to do complex tasks, since describing them takes almost as long as implementing.

    Published Dec 14, 2025self-described: Team working in a monorepo of over 1M lines of code, moved from Copilot to Windsurf

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-12-14
    Counts toward
    Delivered outcomes, context handling: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Still applicable
    Observed on
    Windsurf, Supercomplete

  30. Positive

    Tripti Sheth, Senior Director of Software Engineering at Salesforce

    Building AI-Powered Migration Tools: Compressing 4 Sprints Into 3 Days with Cursor, Windsurf, and ClaudeCompany engineering blog post, Salesforce Engineering

    Sheth's backend team built a migration tracking UI with Cursor, Windsurf and Claude in three days, work that typically takes about six weeks; Windsurf's rules and memories let it refactor unwieldy code into modular structures.

    "Windsurf: We used rules and memories, allowing it to learn from previous context and refactor code into more manageable structures."
    Excerpt, Tripti Sheth

    Published Nov 17, 2025Monitoring Cloud Notification Management team; backend engineers building a migration tracking tool

    Practitioner account, name and role on their organization's own page; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Name and role on their organization's own page; Used it directly
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (Senior Director of Software Engineering)
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-11-17
    Counts toward
    Delivered outcomes, correctness: favorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, context handling: favorable, G1 (1 unit before bounds), capped: C-relationship

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf rules and memories, beside Cursor and Claude

  31. Mixed

    CSMastermind, Hacker News commenter who evaluates dev tooling for their company

    Comment on: Codemaps: Understand Code, Before You Vibe ItHacker News comment

    Still rates Windsurf ahead of Cursor but has moved to Codex for its cloud environments; pricing kept them from rolling Windsurf out across their company.

    Published Nov 5, 2025self-described: Decides tooling across their company

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-11-05
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Needs a recheck

  32. Positive

    bluelightning2k, Hacker News commenter, senior engineer

    Comment on: Codemaps: Understand Code, Before You Vibe ItHacker News comment

    A senior engineer and 12-month user, the commenter says developers overlook Windsurf now that debate centers on Cursor, Claude Code and Codex, and praises Codemaps.

    Published Nov 4, 2025self-described: Senior engineer mixing agentic and regular coding

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-11-04
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  33. Positive

    mpalmer, Hacker News commenter

    Comment on: Codemaps: Understand Code, Before You Vibe ItHacker News comment

    Using Windsurf at work alongside Claude Code, the commenter says its UX adds real value and Codemaps have been excellent for weeks, though code churn could make them harder to maintain.

    Published Nov 4, 2025self-described: Uses Windsurf at work alongside Claude Code

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-11-04
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  34. Positive

    Pat Casey, CTO and co-founder of ServiceNow

    ServiceNow Says Windsurf Gave Its Engineers a 10% Productivity BoostPodcast episode page, The New Stack Agents

    ServiceNow evaluated newer coding tools, chose Windsurf over GitHub Copilot and rolled it out to about 7,000 engineers. It then measured about a 10% gain in stories per unit time per engineer from the release before to the release after.

    "…the team evaluated the new crop of AI coding tools and decided to invest in Windsurf, which it has now rolled out to about 7,000 engineers. ServiceNow saw about a 10% increase in productivity as measured by stories per unit time per engineer between the last release before the team started using Windsurf and the first one after the tool was introduced."
    Excerpt, Frederic Lardinois
    "ServiceNow saw about a 10% increase in productivity as measured by stories per unit time per engineer between the last release before the team started using Windsurf and the first one after the tool was introduced."
    Excerpt, The New Stack

    Published Oct 17, 2025Enterprise software company; about 7,000 engineers moved from GitHub Copilot

    not counted until substantiated How GILD checked it
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (Pat Casey, CTO of ServiceNow)
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Reports others' experience
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-10-17, needs a recheck
    Not counted
    Delivered outcomes, measured outcomes: a lead (G0), counts toward nothing until substantiated

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Podcasts, Talks & Videos
    Who
    Organization-wide
    Evidence
    Measured
    Applies to
    Still applicable
    Observed on
    Company-wide rollout to about 7,000 engineers

  35. Mixed

    joyboy123456, Windsurf user filing a bug report

    [Windsurf Bug] AI keeps getting stuck executing terminal commandsGitHub issue

    Reports that every terminal command the agent ran on Windows PowerShell stayed in a running state and never returned, while the same commands worked when run by hand.

    Published Oct 10, 2025self-described: Solo, Windows with PowerShell

    Practitioner account, pseudonymous; direct use; counts toward Reliability How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-10-10, needs a recheck
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windows, PowerShell

  36. Mixed

    tartieret, Hacker News commenter, CTO

    Comment on: Vibe engineeringHacker News comment

    As CTO the commenter has junior developers run an AI first-pass review before human review to cut pull request iterations; their team used Windsurf, then switched to GitHub Copilot.

    Published Oct 9, 2025self-described: Their engineering team, with senior and junior developers

    Practitioner account, pseudonymous; ran the rollout; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Ran the rollout; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Ran the rollout
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-10-09, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Still applicable

  37. Positive

    Kristin Foss, VP of Software Engineering for Customer Centric Engineering at Salesforce

    How AI Tools Cut Customer Escalation Time: From Days of Manual Work to MinutesCompany engineering blog post, Salesforce Engineering

    Foss's escalation engineers use Cursor, Windsurf and Claude CLI to build working reproductions of customer problems from internal source code, finding issues in large codebases in minutes where setup once took hours or days.

    "To streamline this process, our CCE engineers now leverage Cursor, Windsurf, and Claude CLI."
    Excerpt, Kristin Foss

    Published Sep 29, 2025180-person engineering team resolving customer escalations across Salesforce products

    Practitioner account, name and role on their organization's own page; ran the rollout; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Name and role on their organization's own page; Customer of Devin Desktop; Ran the rollout; Artifact checked (organization page); Needs recheck after 12 months
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (VP of Software Engineering for Customer Centric Engineering (CCE))
    Relationship
    customer (self disclosed); customer (self disclosed); customer (self disclosed)
    How they know
    Ran the rollout
    Support
    3 artifacts (organization page)
    Date and version
    2025-09-29, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, context handling: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: favorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Engineering Blogs & Reviews
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Still applicable
    Observed on
    Windsurf beside Cursor and Claude CLI for customer issue reproductions

  38. Positive

    Bap (fernandezbaptiste), developer

    IDE Comparison with Cursor, Windsurf & Copilot on GPT-5Personal blog post, DEV Community

    Testing all three on GPT-5 over a weekend, the author found Windsurf explained an existing codebase best, kept context across the session and scaffolded a project unprompted, with slower inline suggestions.

    Published Sep 3, 2025self-described: Solo weekend test on greenfield and brownfield projects

    Practitioner account, self-described; structured test; counts toward Delivered outcomes, Reliability, Operational control, Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Ran a structured test; Needs recheck after 12 months; Specific details
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Ran a structured test
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-09-03, GPT-5, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Reliability, run stability: favorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Operational control, execution control: mixed, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Adoption and upkeep, continuity: mixed, G1 (1 unit before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Operational control, admin control: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    GPT-5 in each editor, Aug 22 to 24, 2025

  39. Positive

    scotty79, Hacker News commenter

    Comment on: Why does AI feel so different?Hacker News comment

    Over three days Windsurf wrote multi-file game logic, refactored on request and fixed bugs that would take them half an hour; the commenter made 95% of changes by instructing it.

    Published Aug 13, 2025self-described: Solo, a game project

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-08-13, needs a recheck
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  40. Negative

    cactca, Hacker News commenter and self-described Windsurf subscriber

    Comment on Hacker News: cancelled Windsurf after the acquisitionHacker News comment

    After Cognition acquired Windsurf, this subscriber cancelled and moved to Zed and Claude Code.

    Published Aug 5, 2025self-described: Solo developer

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current

  41. Positive

    Chizaram Ken, frontend developer writing for LogRocket

    Windsurf vs. Cursor: When to choose the challengerIndependent review, LogRocket Blog

    Building a Next.js todo app with Gemini 2.5 Pro in Cascade, Ken saw it update the correct files without being told and import Lucide-react without installing it. After a planned theme change, the theming "seems to have broken somewhere" until a few more iterations fixed it.

    Published Jul 31, 2025self-described: Solo, a Next.js to-do app built with Cascade

    Practitioner account, self-described; direct use; counts toward Delivered outcomes, Operational control, Cost predictability and limits, Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-31, Gemini 2.5 Pro, needs a recheck
    Counts toward
    Delivered outcomes, correctness: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Delivered outcomes, task completion: mixed, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Delivered outcomes, context handling: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Cost predictability and limits, cost transparency: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Adoption and upkeep, setup: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Adoption and upkeep, continuity: mixed, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Operational control, execution control: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Gemini 2.5 Pro in Cascade

  42. Negative

    badaigent, Hacker News commenter

    Comment on: Vibe coding service Replit deleted production database, faked data, told fibsHacker News comment

    Reports that Windsurf's agent replaced their dashboard with its own component, removed files without approval despite their rules, and broke the app again after restoring it.

    Published Jul 27, 2025self-described: Solo, a Next.js app

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes, Reliability, Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Artifact checked (transcript); Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    4 artifacts (transcript)
    Date and version
    2025-07-27, needs a recheck
    Counts toward
    Operational control, execution control: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Reliability, run stability: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, task completion: unfavorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  43. Positive

    Steve Lewis, Lead Software Engineer at World Wide Technology

    Windsurf In Action: Building AWS InfrastructureCompany blog post, WWT

    Lewis reports that the team has used Windsurf for a few months. Planning the architecture with it, Lewis built S3 and CloudFront hosting in Terraform plus a Node.js upload script in less than 1 day against an estimated 2-3 days, and argues such tools need someone to drive them and provide context.

    "Our team has been using Windsurf as an AI-powered development environment for a few months now."
    Excerpt, Steve Lewis

    Published Jul 23, 2025Team building a mobile app backend on AWS with Terraform; WWT also sells Windsurf adoption services

    Practitioner account, name and role on their organization's own page; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Name and role on their organization's own page; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (Lead Software Engineer)
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-23, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf on Terraform, CloudFront and a Node.js upload script

  44. Mixed

    justinclift, Hacker News commenter

    Comment on: Cognition (Devin AI) to Acquire WindsurfHacker News comment

    Their workplace still uses Windsurf, but the commenter canceled their personal plan because Claude Sonnet 3.7 output and issue analysis in it were too often wrong to justify paying.

    Published Jul 16, 2025self-described: Their workplace uses Windsurf; the commenter canceled a personal plan

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-16, Claude Sonnet 3.7 (thinking), needs a recheck
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Claude Sonnet 3.7 (thinking)

  45. Positive

    David Roff, Software Engineer at World Wide Technology

    How Windsurf Has Revolutionized My Development WorkflowCompany blog post, WWT

    Roff reports a personal productivity gain above 70% and similar gains across his engineering team, to the point that QA struggles to keep up, while every Windsurf change still goes through a senior engineer's review.

    "Team-wide benefits: The entire engineering team has experienced similar productivity gains."
    Excerpt, David Roff

    Published Jul 14, 2025Full stack platform team: web, React Native mobile, APIs and databases; WWT also sells Windsurf adoption services

    Practitioner account, name and role on their organization's own page; direct use; counts toward Delivered outcomes, Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Name and role on their organization's own page; Used it directly; Artifact checked (organization page); Needs recheck after 12 months
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (Software Engineer - 4)
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    4 artifacts (organization page)
    Date and version
    2025-07-14, Claude 3.7 Sonnet, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, context handling: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Claude 3.7 Sonnet and o3 in Windsurf

  46. Positive

    RestlessAPI, Hacker News commenter

    Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleHacker News comment

    Uses Windsurf mainly for completions that predict their next edit, keeping them in control of the codebase and writing everything themselves without delegating to an agent.

    Published Jul 12, 2025self-described: Solo; completion-first use

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-12, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Still applicable

  47. Negative

    chris_engel, Hacker News commenter

    Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleHacker News comment

    Tried agent workflows in Copilot and Windsurf on Claude 3.5 and 4; they got stuck or built themselves into corners, and the output was fragile enough to refactor immediately.

    Published Jul 12, 2025self-described: Solo; tried agent workflows in Copilot and Windsurf

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-12, claude 3.5 and 4, needs a recheck
    Not counted
    Reliability, run stability: repeats a stronger account in its group
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Claude 3.5 and Claude 4

  48. Mixed

    sunaookami, Hacker News commenter

    Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleHacker News comment

    Saw Windsurf as slightly better than Cursor, with a nicer chat panel and autocomplete and a US$5 cheaper plan, until losing Anthropic model access cost it users.

    Published Jul 11, 2025self-described: Former Windsurf user

    Practitioner account, pseudonymous; opinion; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Opinion; Specific details; Needs recheck after a product change
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Opinion, no use stated
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-11, needs a recheck
    Not counted
    Cost predictability and limits, cost transparency: repeats a stronger account in its group
    Not counted
    Adoption and upkeep, continuity: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  49. Negative

    diegof79, Hacker News commenter

    Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleHacker News comment

    Adopted Windsurf at launch for its UX, then saw more issues with each update, including the agent hanging on shell commands, and moved back to GitHub Copilot.

    Published Jul 11, 2025self-described: Solo; adopted Windsurf at launch

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-07-11, needs a recheck
    Not counted
    Reliability, regressions: repeats a stronger account in its group
    Not counted
    Reliability, run stability: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  50. Mixed

    Kieran Klaassen, general manager of Cora at Every

    Transcript: 'How Two Engineers Ship Like a Team of 15 With AI Agents'Podcast episode transcript, AI & I (Every)

    Klaassen ranks Windsurf a C because it lacks Claude 4, though three weeks earlier it would have been an A, and had switched from Cursor to Windsurf a few months back before switching back. Calls the Cursor and Windsurf way of coding great, one step before agents like Claude Code.

    "So we need to rethink again, hey, Cursor, Windsurf, the old school way of coding was great."
    Excerpt, Kieran Klaassen

    Published Jun 11, 2025Two-engineer team building Cora, an AI email product at Every

    Practitioner account, name and role on their organization's own page; direct use; counts toward Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Name and role on their organization's own page; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Name on a page the person does not control; employer on a page the person does not control; role on a page the person does not control (Member of Frontier Team)
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-06-11, needs a recheck
    Counts toward
    Operational control, models and deployment: unfavorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Podcasts, Talks & Videos
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Still applicable
    Observed on
    Cursor and Windsurf before a move to Claude Code

  51. Mixed

    Mark Winteringham, tester, toolsmith, author and instructor

    Lessons in quality engineering from working with Cursor and WindsurfArticle, Ministry of Testing

    Working in small tasks after a messy start in Cursor, Winteringham had the Windsurf agent fix each failing unit test quickly, though it wrote more code than he expected.

    "After that, I worked with the Windsurf agent to fix each of the failing tests. This process worked well, and it didn’t take too much time to get them all to pass."
    Excerpt, Mark Winteringham

    Published Jun 10, 2025self-described: Solo; migrating his public restful-booker-platform front end to Next.js

    Practitioner account; direct use; counts toward Delivered outcomes, Reliability, Cost predictability and limits How GILD checked it
    Origin
    Published practitioner account
    Checked
    Used it directly; Needs recheck after 12 months
    Role
    Name unknown; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-06-10, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G1 (1 unit before bounds), capped: C-relationship, C-unknown-control
    Counts toward
    Delivered outcomes, correctness: unfavorable, G1 (1 unit before bounds), capped: C-relationship, C-unknown-control
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relationship, C-unknown-control
    Counts toward
    Cost predictability and limits, limits in practice: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship, C-unknown-control

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf trial and free agent

  52. Mixed

    osigurdson, Hacker News commenter

    Comment on: My AI skeptic friends are all nutsHacker News comment

    After a few months the commenter finds Windsurf useful when tired or in unfamiliar areas, but waiting and errors break flow. On familiar work the commenter puts it at half the speed of typing.

    Published Jun 3, 2025self-described: Solo, a few months of use

    Practitioner account, pseudonymous; direct use; counts toward Reliability How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-06-03, needs a recheck
    Counts toward
    Reliability, service availability: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  53. No opinion stated

    Maxwell Zeff, senior reporter, TechCrunch

    TechCrunch: Windsurf says Anthropic is limiting its direct access to Claude AI modelsNews report

    Anthropic cut Windsurf's direct Claude capacity with little notice in June 2025, and a Windsurf customer since late 2024 switched to Cursor to keep using Claude 4.

    "Windsurf CEO Varun Mohan said in a post on X Tuesday that Anthropic gave Windsurf little notice for this change, and the startup now has to find other third-party compute providers to run several of Anthropic’s most popular AI models on its platform."
    Excerpt, TechCrunch

    Published Jun 3, 2025self-described: Windsurf subscribers using Claude models

    Counts toward

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Current

  54. Mixed

    tianshuo, Hacker News commenter

    Comment on: Windsurf SWE-1: Our First Frontier ModelsHacker News comment

    A paid user, the commenter asks Windsurf to stop chasing frontier models and build predictable workflows: task boards, TDD, git branches without reliance on undo, and docs kept in sync with code.

    Published May 19, 2025self-described: Paid individual user

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-19, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  55. Negative

    robch, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Using their company's Windsurf license for staff-level Android tooling work, the commenter finds autocomplete distracting and the agent hallucinating answers for uncommon tasks.

    Published May 18, 2025self-described: Staff-level Android tooling work on a company license

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-18, needs a recheck
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Professional team
    Evidence
    Reported
    Applies to
    Needs a recheck

  56. Positive

    eisfresser, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Chooses Windsurf, running parallel Cascade sessions, one investigating a bug while another builds a feature, and drops in later to review and accept the changes.

    Published May 12, 2025self-described: Solo, parallel Cascade sessions

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  57. Negative

    pbowyer, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Trialling both editors on the same PHP codebases, the commenter got poorer results from Windsurf on existing code, saw it read files 50 lines at a time, and went back to PhpStorm.

    Published May 12, 2025self-described: Trialled Cursor and Windsurf on the same PHP codebases

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Counts toward
    Delivered outcomes, context handling: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  58. Mixed

    pembrook, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Once ahead of Cursor on agentic coding in their view, Windsurf now frustrates them by gathering context 100 to 200 lines at a time. The commenter blames that for bugs and poor results.

    Published May 12, 2025self-described: Solo, agentic coding

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Counts toward
    Delivered outcomes, context handling: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  59. Negative

    powerapple, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Testing Windsurf, the commenter found it installed dependencies into their global Python without a virtual environment and rewrote code the commenter had not asked it to change.

    Published May 12, 2025self-described: Solo, a Python project

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  60. Positive

    whywhywhywhy, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Uses Cursor until hitting a wall, then switches to Windsurf, which the commenter says solves issues Cursor fails on after 40 minutes of retries, though its plans are limited.

    Published May 12, 2025self-described: Solo; uses Cursor and Windsurf

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, unlimited “cascade base”, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Cost predictability and limits, limits in practice: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  61. Negative

    artdigital, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Found the same models produced noticeably worse code in Windsurf than in Zed, Copilot or Cursor, with erratic tool use, so the commenter canceled and made Copilot their daily driver.

    Published May 12, 2025self-described: Solo; former subscriber

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group
    Not counted
    Reliability, run stability: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  62. Positive

    thesurlydev, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    Roughs in the outline themselves, then has the Cascade plugin in JetBrains add tests, performance improvements and idiomatic cleanups; the commenter plans to add global and project rules.

    Published May 12, 2025self-described: Uses the Windsurf plugin inside JetBrains IDEs

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf JetBrains plugin with Cascade

  63. Mixed

    ksymph, Hacker News commenter

    Comment on: Ask HN: Cursor or Windsurf?Hacker News comment

    After switching between both, the commenter prefers Cursor's completions, credits Windsurf with better project awareness across files that is hard to trigger, and notes its lower price.

    Published May 12, 2025self-described: Solo; switches between Cursor and Windsurf

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-12, needs a recheck
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Cost predictability and limits, cost transparency: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  64. Mixed

    raymondgh, Hacker News commenter

    Comment on: Ask HN: How much better are AI IDEs vs. copy pasting into chat apps?Hacker News comment

    Vibe coding a web app, the commenter found Cascade good at locating where to change the codebase, but it could not solve performance or CSS animation problems without their direction.

    Published May 8, 2025self-described: Solo, a web app built with Windsurf

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-08, needs a recheck
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  65. Mixed

    sagarpatil, Hacker News commenter

    Comment on: OpenAI reaches agreement to buy Windsurf for US$3BHacker News comment

    A customer since launch, the commenter handed Windsurf a broken Alembic migration after an hour of manual debugging; it isolated the file, fixed it, added tests and opened a PR. The commenter now prefers Claude Code.

    Published May 7, 2025self-described: Solo; Windsurf customer since launch

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-07, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  66. Positive

    johntarter, Hacker News commenter

    Comment on: OpenAI reaches agreement to buy Windsurf for US$3BHacker News comment

    Moving between Windsurf and GitHub Copilot, the commenter credits Windsurf's whole-codebase indexing for completions across files and Cascade's terminal handling, while rating Cursor's checkpoint undo higher.

    Published May 7, 2025self-described: Evaluating Windsurf against GitHub Copilot

    Practitioner account, pseudonymous; direct use; counts toward Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-07, needs a recheck
    Counts toward
    Adoption and upkeep, continuity: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Operational control, execution control: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Still applicable

  67. Negative

    zero-g, Hacker News commenter

    Comment on: OpenAI reaches agreement to buy Windsurf for US$3BHacker News comment

    Each time the commenter tried the Windsurf Editor and its Vim and IntelliJ plugins it felt unpolished, and autocomplete felt much snappier in Cursor and in GitHub Copilot.

    Published May 6, 2025self-described: Tried the editor and the Vim and IntelliJ plugins

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-06, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  68. Positive

    __jl__, Hacker News commenter

    Comment on: OpenAI reaches agreement to buy Windsurf for US$3BHacker News comment

    Comparing approaches, the commenter observes that Windsurf still loads more code into context than Cursor's search-tool approach, giving it an edge in some cases at a lower subscription price.

    Published May 6, 2025self-described: Uses Cursor and Windsurf

    Practitioner account, pseudonymous; opinion; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Opinion; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Opinion, no use stated
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-06, needs a recheck
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  69. Mixed

    osigurdson, Hacker News commenter

    Comment on: OpenAI reaches agreement to buy Windsurf for US$3BHacker News comment

    The commenter likes Windsurf for exploration but writes code they want to keep by hand, warning that without deep review of every Windsurf change you lose understanding of the codebase.

    Published May 6, 2025self-described: Solo; writes code the commenter keeps by hand in Vim

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-05-06, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Still applicable

  70. Positive

    tartieret, Hacker News commenter (a CTO, by their own account elsewhere)

    Comment on: An update to our pricingHacker News comment

    Their company standardized on Windsurf after reviewing Cursor, because Windsurf handled dev containers and WSL development, which most of its developers use, with fewer rough edges.

    Published Apr 22, 2025self-described: Their company, whose developers mostly work in dev containers and WSL

    Practitioner account, pseudonymous; ran the rollout; counts toward Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Ran the rollout; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Ran the rollout
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-22, needs a recheck
    Counts toward
    Adoption and upkeep, setup: favorable, G1 (1 unit before bounds), capped: C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward

    Published on
    Developer Communities
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Still applicable

  71. Mixed

    CSMastermind, Hacker News commenter who evaluates dev tooling for their teams

    Comment on: An update to our pricingHacker News comment

    Evaluating tools for their teams, the commenter found Windsurf the one tool that raised their productivity beyond chat, especially on command-line work, but held back a team rollout over pricing.

    Published Apr 22, 2025self-described: Evaluates tools for their engineering teams

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-22, needs a recheck
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Needs a recheck

  72. Positive

    jawns, Hacker News commenter

    Comment on: An update to our pricingHacker News comment

    As an avid Windsurf user, supports the simplification of the pricing: in nearly all cases the author doesn't care how many steps the model takes, and just wants it to do what was asked.

    Published Apr 21, 2025self-described: Individual Windsurf subscriber

    Practitioner account, pseudonymous; direct use; not counted, earlier release How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Out of date (earlier release)
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-21, out of date
    Not counted
    no aspect: excluded, describes an earlier release or was retracted; takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward nothing current

    Replaced by newer evidence. Kept in the source history with its scope stated; it takes part in no current figure.

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Out of date
    Observed on
    April 2025 pricing change

  73. Mixed

    SquareWheel, Hacker News commenter

    Comment on: An update to our pricingHacker News comment

    In their first week completions helped with repetitive renames across files but often mangled unrelated code, the UI felt noisy, and chat sometimes scanned files endlessly with no output.

    Published Apr 21, 2025self-described: First week of individual use

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-21, unlimited GPT-4.1 model, needs a recheck
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group
    Not counted
    Delivered outcomes, correctness: repeats a stronger account in its group
    Not counted
    Reliability, run stability: repeats a stronger account in its group
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    GPT-4.1 during its free period

  74. Negative

    ramesh31, Hacker News commenter

    Comment on: An update to our pricingHacker News comment

    A daily full-time Windsurf user, the commenter calls it inferior to open-source agents such as Cline and Claude Code that use provider keys directly, without flow credits.

    Published Apr 21, 2025self-described: Daily full-time use

    Practitioner account, pseudonymous; direct use; not counted, repeats a stronger account How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-21, needs a recheck
    Not counted
    Operational control, models and deployment: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Flow-credit pricing

  75. Mixed

    adamgordonbell, Hacker News commenter

    Comment on: An update to our pricingHacker News comment

    On the largest Windsurf plan the commenter still tops up 10 dollars at a time, which works out to paying per task; the commenter finds the plans undersized for professional work.

    Published Apr 21, 2025self-described: Individual, professional work

    Practitioner account, pseudonymous; direct use; not counted, earlier release How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Specific details; Out of date (the largest plan)
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-21, the largest plan, out of date
    Not counted
    Cost predictability and limits, limits in practice: excluded, describes an earlier release or was retracted
    Not counted
    Cost predictability and limits, cost transparency: excluded, describes an earlier release or was retracted

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward nothing current

    Replaced by newer evidence. Kept in the source history with its scope stated; it takes part in no current figure.

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Out of date
    Observed on
    Largest individual plan with credit top-ups

  76. Positive

    AndrewKemendo, Hacker News commenter writing as a tech founder planning team size

    Comment on: The Bitter PredictionHacker News comment

    The commenter and a fellow founder made Windsurf with Claude their default environment and plan future projects with about a third fewer engineers.

    Published Apr 13, 2025self-described: Tech founder sizing engineering teams for future projects

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-04-13, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Published on
    Developer Communities
    Who
    Technical leadership
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf with Claude

  77. Positive

    Matt Wolfe, founder of FutureTools and technology video creator

    Matt Wolfe on X: tips after 100+ hours of vibe codingX post

    Wolfe's first tip for vibe coding is to switch from Cursor to Windsurf, which he finds much better at remembering past context.

    "Switch to Windsurf from Cursor. Windsurf seems much better at remembering past context for some reason."
    Excerpt, Matt Wolfe

    Published Mar 25, 2025self-described: Solo; more than 100 hours of vibe coding in Cursor and Windsurf

    Practitioner account, pseudonymous; direct use; counts toward Delivered outcomes, Reliability, Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-03-25, needs a recheck
    Counts toward
    Delivered outcomes, context handling: favorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Operational control, execution control: mixed, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  78. Positive

    John Hoestje, Software Architect and Developer at Keyhole Software

    Windsurf’s Game-Changing IDE Experience, Part 2, Coding ExamplesCompany blog post, Keyhole Software Dev Blog

    Hoestje built two apps from Cascade prompts without editing a line by hand, and says he fell short when he moved too fast to review Cascade's suggested changes.

    "I fell short when I was trying to go too fast. I didn’t take the time to properly review the suggested changes by Cascade."
    Excerpt, John Hoestje

    Published Mar 20, 2025Software consultancy; solo experiments building a React site and a Flutter Sudoku game

    Practitioner account, self-described; direct use; counts toward Delivered outcomes, Reliability, Operational control, Adoption and upkeep How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Needs recheck after 12 months; Specific details
    Role
    Name self-described; employer self-described; role self-described (Software Architect and Developer)
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-03-20, needs a recheck
    Counts toward
    Delivered outcomes, task completion: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Delivered outcomes, correctness: mixed, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relationship
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Adoption and upkeep, setup: favorable, G1 (1 unit before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Operational control, execution control: repeats a stronger account in its group
    Not counted
    Operational control, data control: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Codeium Windsurf with Cascade

  79. Mixed

    AldrinSalazar, Windsurf user filing a bug report

    Cascade stuck in a command execution loop because it cannot execute commandsGitHub issue

    Reports that on Windows 11 every Cascade attempt to run a command failed and left the agent stuck in a loop.

    Published Mar 16, 2025self-described: Solo, Windows 11

    Practitioner account, pseudonymous; direct use; counts toward Reliability How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    1 artifact (screenshot)
    Date and version
    2025-03-16, Windsurf Version: 1.4.6, needs a recheck
    Counts toward
    Reliability, run stability: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 30, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf 1.4.6 on Windows 11

  80. Positive

    Søren Kottal, Umbraco developer

    Vibe coding a nifty Umbraco toolPersonal blog post, DEV Community

    Built an Umbraco thumbnail tool with Windsurf in a few hours, but it reintroduced old bugs two or three times. The author did not read the code, so would not use it for supported work.

    Published Mar 16, 2025self-described: Solo, a small web tool built in an afternoon

    Practitioner account, self-described; direct use; counts toward Delivered outcomes How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-03-16, needs a recheck
    Counts toward
    Delivered outcomes, correctness: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck

  81. Positive

    divan, Hacker News commenter

    Comment on: I use Cursor daily - here's how I avoid the garbage partsHacker News comment

    Switched from Cursor to Windsurf because it used far fewer credits; on a US$10 monthly plan the commenter could not reach the limits despite daily use.

    Published Mar 12, 2025self-described: Daily individual use

    Practitioner account, pseudonymous; direct use; not counted, earlier release How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Out of date (earlier release); Specific details
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-03-12, out of date
    Not counted
    Cost predictability and limits, limits in practice: excluded, describes an earlier release or was retracted
    Not counted
    Cost predictability and limits, cost transparency: excluded, describes an earlier release or was retracted

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward nothing current

    Replaced by newer evidence. Kept in the source history with its scope stated; it takes part in no current figure.

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Out of date
    Observed on
    Credit-based plan, US$10 per month

  82. Positive

    Daniel Bergholz, developer

    My AI-Powered Workflow for Writing Elixir and Phoenix with WindsurfPersonal blog post, DEV Community

    After a month with each, Bergholz chose Windsurf for better codebase understanding and a smoother UI, and uses memories and a .windsurfrules file to correct it on a niche Elixir stack.

    Published Mar 10, 2025self-described: Solo side projects in Elixir, Phoenix, React and Inertia

    Practitioner account, self-described; direct use; counts toward Delivered outcomes, Cost predictability and limits How GILD checked it
    Origin
    Published practitioner account
    Checked
    Self-described; Customer of Devin Desktop; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Name self-described; employer unknown; role unknown
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-03-10, Claude 3.5 Sonnet, needs a recheck
    Counts toward
    Delivered outcomes, correctness: mixed, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Cost predictability and limits, cost predictability: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group
    Not counted
    Delivered outcomes, task completion: repeats a stronger account in its group

    Checked Oct 31, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Engineering Blogs & Reviews
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Claude 3.5 Sonnet

  83. Positive

    bicx, Hacker News commenter

    Comment on: Claude 3.7 Sonnet and Claude CodeHacker News comment

    Recommends Cascade for agentic code writing and exploration, saying it saved them many hours understanding new codebases and tracing data flows; Claude 3.5 cost credits.

    Published Feb 24, 2025self-described: Individual use on new codebases

    Practitioner account, pseudonymous; direct use; counts toward Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-02-24, Claude 3.5, needs a recheck
    Counts toward
    Operational control, models and deployment: favorable, G2 (2 units before bounds), capped: C-relationship
    Not counted
    Delivered outcomes, context handling: repeats a stronger account in its group

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Claude 3.5 Sonnet and DeepSeek V3 in Cascade

  84. Mixed

    nadavoid, Windsurf user filing a feature request

    Remove or increase limit of past conversations in Cascade (Windsurf)GitHub issue

    Asks Windsurf to lift Cascade's limit of 20 saved conversations, since the oldest is lost for good when a 21st starts and past conversations are worth revisiting.

    Published Feb 18, 2025self-described: Solo, Cascade conversations

    Practitioner account, pseudonymous; direct use; counts toward Operational control How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Specific details; Needs recheck after a product change
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-02-18, needs a recheck
    Counts toward
    Operational control, data control: unfavorable, G1 (1 unit before bounds), capped: C-relevance, C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Cascade conversation history

  85. Positive

    thesurlydev, Hacker News commenter

    Comment on: JetBrains Fleet drops support for Kotlin MultiplatformHacker News comment

    Runs Windsurf next to JetBrains IDEs because the commenter finds Cascade ahead of the 30-plus LLM plugins available for JetBrains, though juggling two editors is a pain.

    Published Feb 12, 2025self-described: Uses JetBrains IDEs with Windsurf alongside

    Practitioner account, pseudonymous; direct use; not counted, takes no position How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Used it directly; Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    Checked only from the record itself; an outside check is open
    How they know
    Used it directly
    Support
    No artifact or independent corroboration recorded
    Date and version
    2025-02-12, needs a recheck
    Not counted
    no aspect: excluded, takes no position on any assessed aspect

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf with Cascade

  86. Mixed

    Bwogi, Windsurf subscriber filing a bug report

    The model produced an invalid tool call, trying again.GitHub issue on Exafunction/codeium

    After an update, every Cascade step failed with an invalid tool call and then an internal error, leaving the reporter unable to use a US$60 monthly subscription that had worked smoothly before.

    "…i was coding smoothly without any issues before the update. Now i cannot use my 60 buck monthly subscription anymore."
    Excerpt, Bwogi

    Published Jan 30, 2025self-described: Solo; Next.js, TypeScript and Tailwind on macOS

    Practitioner account, pseudonymous; direct use; counts toward Reliability How GILD checked it
    Origin
    Published practitioner account
    Checked
    Pseudonymous; Customer of Devin Desktop; Used it directly; Artifact checked (log); Needs recheck after 12 months
    Role
    Pseudonymous account; identity not pursued
    Relationship
    customer (self disclosed)
    How they know
    Used it directly
    Support
    1 artifact (log)
    Date and version
    2025-01-30, needs a recheck
    Counts toward
    Reliability, run stability: unfavorable, G2 (2 units before bounds), capped: C-relationship
    Counts toward
    Reliability, regressions: unfavorable, G2 (2 units before bounds), capped: C-relationship

    Checked Oct 29, 2026 under gild-policy/1.0.0. The scorecard

    Counts toward, pending a recheck

    Published on
    Developer Communities
    Who
    Individual practitioner
    Evidence
    Reported
    Applies to
    Needs a recheck
    Observed on
    Windsurf after a January 2025 update

Product documentation 20

Pages from Cognition, used to check facts for every group.

  1. Devin Desktop FAQ

    States that Windsurf became Devin Desktop on June 2, 2026 with plans and pricing unchanged, that the app is now named Devin, and that the Windsurf JetBrains plugin is in maintenance mode.

    Checked Oct 12, 2026

  2. Devin Local agent

    Documents Devin Local as the default agent for new tabs, with allow, ask, and deny permissions, sandboxing, subagents, and up to 30% fewer tokens than Cascade; memories and workflows are not supported.

    Checked Oct 12, 2026

  3. Cascade overview

    Describes Cascade as one of two local agents in Devin Desktop, alongside Devin Local.

    Checked Oct 12, 2026

  4. Quota-based usage

    Explains the March 2026 move from credits to daily and weekly quotas drawn down per token, with extra usage billed at API list prices and SSO now Enterprise only.

    Checked Oct 12, 2026

  5. Self-serve plans

    Lists Pro at US$20 and Max at US$200 per month, and Teams at US$40 per full seat with a US$80 monthly minimum; flex seats do not include Devin Desktop.

    Checked Oct 12, 2026

  6. Plans and usage

    Describes the Free, Pro, Max, Teams, and Enterprise plans, and Enterprise billing in Agent Compute Units for newer contracts.

    Checked Oct 12, 2026

  7. Admin controls reference

    Lists every setting admins can enforce: permission rules, sandbox enforcement, MCP and model allowlists, data-retention models, and device policies.

    Checked Oct 12, 2026

  8. Terminal

    Documents Cascade's auto-execution levels and allow and deny lists, and the admin-set maximum level on Teams and Enterprise.

    Checked Oct 12, 2026

  9. Memories and rules

    Explains rules in .devin/rules, with .windsurf/rules as a fallback, and that memories apply only to the legacy Cascade agent.

    Checked Oct 12, 2026

  10. AGENTS.md

    Describes AGENTS.md files as rules scoped by directory, read by both Cascade and Devin Local.

    Checked Oct 12, 2026

  11. Context awareness

    Describes the context engine that indexes a codebase for suggestions and agent context.

    Checked Oct 12, 2026

  12. .devinignore

    Documents excluding paths from indexing and agent access with .devinignore, in addition to .gitignore.

    Checked Oct 12, 2026

  13. Agent Command Center

    Describes the board that tracks local and cloud agent sessions by status.

    Checked Oct 12, 2026

  14. Devin in Devin Desktop

    Documents delegating work to Devin cloud sessions, each on its own VM, funded from the same quota on self-serve plans.

    Checked Oct 12, 2026

  15. Agent Client Protocol

    Documents running third-party agents such as Codex CLI and Claude Agent inside Devin Desktop, billed directly by their providers.

    Checked Oct 12, 2026

  16. Quick Review

    Documents an agent review of local changes, available only with Devin Local, with the SWE-check model free on every plan.

    Checked Oct 12, 2026

  17. Role-based access

    States that role-based access controls are available on Enterprise plans only.

    Checked Oct 12, 2026

  18. SSO and SCIM

    States that SSO is available to Enterprise users only, and documents SAML single sign-on and SCIM provisioning with Okta, Microsoft Entra ID, Google Workspace, and other identity providers.

    Checked Oct 12, 2026

  19. Guide for admins

    Gives an enterprise rollout checklist: SSO, SCIM, roles, analytics, and client distribution.

    Checked Oct 12, 2026

  20. Get agent output

    Documents the analytics endpoint that reports lines inserted and deleted by the agent and accepted by the team.

    Checked Oct 12, 2026

Source index

Provenance at a glance

Every source GILD read, with the original address and the context used to count it. Open a finding to filter the full cards above. On smaller screens the index lists each source once by name, cohort and date; its full record above carries the address, context and findings.

Provenance for every analyzed Signals source
SourceOriginal URLAuthor and contextWhoPublishedReview dateAppliesFindings
Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraID dc-62 | Developer Communitiesnews.ycombinator.com/item?id=49652815princevegeta89, Hacker News commenterself-described: Long-time Windsurf userIndividual practitionerPublished Sep 11, 2026Assessment reviewed Nov 1, 2026Current
Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraID dc-63 | Developer Communitiesnews.ycombinator.com/item?id=49653534fishtoaster, Hacker News commenterself-described: Solo; evaluated cloud agents after leaving CursorIndividual practitionerPublished Sep 11, 2026Assessment reviewed Nov 1, 2026Current
Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraID dc-58 | Developer Communitiesnews.ycombinator.com/item?id=49646735bluelightning2k, Hacker News commenterself-described: Uses Devin Desktop as their editorIndividual practitionerPublished Sep 10, 2026Assessment reviewed Nov 1, 2026Current
Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraID dc-59 | Developer Communitiesnews.ycombinator.com/item?id=49647937PolCPP, Hacker News commenterself-described: Grandfathered individual subscriberIndividual practitionerPublished Sep 10, 2026Assessment reviewed Nov 1, 2026Current
Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraID dc-60 | Developer Communitiesnews.ycombinator.com/item?id=49649457hightrix, Hacker News commenterself-described: Their company, on an enterprise plan for about nine monthsProfessional teamPublished Sep 10, 2026Assessment reviewed Nov 1, 2026Current
Comment on: Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-AstraID dc-61 | Developer Communitiesnews.ycombinator.com/item?id=49651044paimapi, Hacker News commenterself-described: Former Windsurf userIndividual practitionerPublished Sep 10, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceID dc-57 | Developer Communitiesnews.ycombinator.com/item?id=48839881RussianCow, Hacker News commenterself-described: Solo, under a month on DevinIndividual practitionerPublished Jul 9, 2026Assessment reviewed Nov 1, 2026Current
Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceID dc-53 | Developer Communitiesnews.ycombinator.com/item?id=48834356fallinditch, Hacker News commenterself-described: Former Windsurf user, now on Devin DesktopIndividual practitionerPublished Jul 8, 2026Assessment reviewed Nov 1, 2026Current
Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceID dc-54 | Developer Communitiesnews.ycombinator.com/item?id=48835982godzillabrennus, Hacker News commenterself-described: Their organization, a former Windsurf customerProfessional teamPublished Jul 8, 2026Assessment reviewed Nov 1, 2026Current
Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceID dc-55 | Developer Communitiesnews.ycombinator.com/item?id=48837314RussianCow, Hacker News commenterself-described: Solo, spec-driven workflowIndividual practitionerPublished Jul 8, 2026Assessment reviewed Nov 1, 2026Current
Comment on: SWE-1.7 Reach Near GPT 5.5 and Opus IntelligenceID dc-56 | Developer Communitiesnews.ycombinator.com/item?id=48838536settled, Hacker News commenterself-described: Solo; used Windsurf before the renameIndividual practitionerPublished Jul 8, 2026Assessment reviewed Nov 1, 2026Current
2x the power users: How structured AI training scaled developer productivityID ld-06 | Podcasts, Talks & Videosgetdx.com/podcast/how-structured-ai-training-scaled-developer-productivity/Jeff Davis, VP of Core Infrastructure at IndeedJob site with more than 2,000 engineers; talk given with product manager Michael ReddingTechnical leadershipPublished Jun 29, 2026Assessment reviewed Nov 1, 2026CurrentListed, no finding position
Mapping the new SDLC at BNY: Codifying AI into every step of the delivery lifecycleID ld-07 | Podcasts, Talks & Videosgetdx.com/podcast/codifying-ai-into-every-step-delivery-lifecycle/Jason Valentino, Head of Software Engineering Strategy at BNYBank with more than 8,000 engineers; owns developer tooling and DevExTechnical leadershipPublished Jun 8, 2026Assessment reviewed Nov 1, 2026Current
Devin Desktop (ex-Windsurf): Features, Pricing, What ChangedID ir-12 | Engineering Blogs & Reviewswww.danilchenko.dev/posts/devin-desktop-review/Maksim Danilchenko, Engineering Lead at inDriveself-described: Hands-on review after the rename; personal blogIndividual practitionerPublished Jun 6, 2026Assessment reviewed Nov 1, 2026Current
From Windsurf to Devin Desktop: my first impressions after migratingID ir-11 | Engineering Blogs & Reviewsdev.to/botoom/from-windsurf-to-devin-desktop-my-first-impressions-after-migrating-1hdbEdwar Diaz, developerself-described: Solo; more than a year on WindsurfIndividual practitionerPublished Jun 4, 2026Assessment reviewed Nov 1, 2026Current
Windsurf Review 2026: Not For Solo Founders, Great For Small TeamsID ir-10 | Engineering Blogs & Reviewsdev.to/rikuq/windsurf-review-2026-not-for-solo-founders-great-for-small-teams-11hoRavi Patel, solo founderself-described: Solo founder; one-week trial on the free tierIndividual practitionerPublished May 25, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Changes to GitHub Copilot individual plansID dc-52 | Developer Communitiesnews.ycombinator.com/item?id=47861910isqueiros, Hacker News commenter at a company with company-wide Windsurf accountsself-described: Company with company-wide Windsurf, Copilot and ChatGPT Enterprise accountsOrganization-widePublished Apr 22, 2026Assessment reviewed Nov 1, 2026Still applicable
Comment on: Anthropic downgraded cache TTL on March 6thID dc-51 | Developer Communitiesnews.ycombinator.com/item?id=47737896zazibar, Hacker News commenter at a company with over 400 engineersself-described: Company of over 400 engineers that canceled its Windsurf subscriptionsOrganization-widePublished Apr 12, 2026Assessment reviewed Nov 1, 2026Still applicable
Comment on: Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClawID dc-50 | Developer Communitiesnews.ycombinator.com/item?id=47640519visarga, Hacker News commenterself-described: Former Windsurf subscriberIndividual practitionerPublished Apr 4, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: US$50 monthly budget, which coding models would you recommend now?ID dc-49 | Developer Communitiesnews.ycombinator.com/item?id=47502399Patt_, Hacker News commenterself-described: Solo on a US$50 monthly tooling budgetIndividual practitionerPublished Mar 24, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Who is still using Windsurf and why?ID dc-48 | Developer Communitiesnews.ycombinator.com/item?id=47458321dennisjoseph, Hacker News commenterself-described: Solo; Windsurf is their primary IDEIndividual practitionerPublished Mar 20, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Windsurf Bug Report - Memory Leak Consuming All RAM on LinuxID dc-47 | Developer Communitiesgithub.com/Exafunction/codeium/issues/289oib, Windsurf user filing a bug reportself-described: Solo, Debian 13 with GNOMEIndividual practitionerPublished Mar 10, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Labor market impacts of AI: A new measure and early evidenceID dc-46 | Developer Communitiesnews.ycombinator.com/item?id=47274141sensanaty, Hacker News commenterself-described: Engineer at a company with AGENTS.md and rules files in placeIndividual practitionerPublished Mar 6, 2026Assessment reviewed Nov 1, 2026Needs a recheck
I Tried Windsurf: An AI Native Coding EnvironmentID ir-08 | Engineering Blogs & Reviewsdev.to/saloniagrawal/i-tried-windsurf-an-ai-native-coding-environment-59i0Saloni Agrawal, React and React Native developerself-described: Solo; VS Code user trying WindsurfIndividual practitionerPublished Mar 4, 2026Assessment reviewed Nov 1, 2026Needs a recheck
# Windsurf IDE Freeze Bug ReportID dc-45 | Developer Communitiesgithub.com/Exafunction/codeium/issues/284aeongeon, Windsurf user filing a bug reportself-described: Solo, Linux (Pop!_OS)Individual practitionerPublished Feb 10, 2026Assessment reviewed Nov 1, 2026Needs a recheck
Latent Space: Brex's AI Hail Mary, with James ReggioID sf-4 | Podcasts, Talks & Videoslatent.space/p/brex/James Reggio, chief technology officer, Brexself-described: Brex, a few seats of several tools chosen by employeesOrganization-widePublished Jan 17, 2026Assessment reviewed Nov 1, 2026CurrentListed, no finding position
C3 AI: Autonomous coding agents beyond developer productivityID sf-2 | Engineering Blogs & Reviewsc3.ai/blog/autonomous-coding-agents-beyond-developer-productivitySina Pakazad, Vice President, Data Science, C3 AIself-described: C3 AI, coding agents inside its platformOrganization-widePublished Dec 24, 2025Assessment reviewed Nov 1, 2026Current
How AI-Enabled Tooling Boosted Code Output 30% While Keeping Quality and Deployment Safety IntactID ld-05 | Engineering Blogs & Reviewsengineering.salesforce.com/how-ai-enabled-tooling-boosted-code-output-30-while-keeping-quality-and-deployment-safety-intact/Darryn Dieken, Chief Availability Officer at SalesforceEngineering productivity and reliability for thousands of Salesforce engineers across 25 Hyperforce regionsTechnical leadershipPublished Dec 16, 2025Assessment reviewed Nov 1, 2026Still applicable
Comment on: Ask HN: How can I get better at using AI for programming?ID dc-44 | Developer Communitiesnews.ycombinator.com/item?id=46266283gcbirzan, Hacker News commenter whose team moved from Copilot to Windsurfself-described: Team working in a monorepo of over 1M lines of code, moved from Copilot to WindsurfProfessional teamPublished Dec 14, 2025Assessment reviewed Nov 1, 2026Still applicable
Building AI-Powered Migration Tools: Compressing 4 Sprints Into 3 Days with Cursor, Windsurf, and ClaudeID ld-01 | Engineering Blogs & Reviewsengineering.salesforce.com/building-ai-powered-migration-tools-compressing-4-sprints-into-3-days-with-cursor-windsurf-and-claude/Tripti Sheth, Senior Director of Software Engineering at SalesforceMonitoring Cloud Notification Management team; backend engineers building a migration tracking toolTechnical leadershipPublished Nov 17, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Codemaps: Understand Code, Before You Vibe ItID dc-43 | Developer Communitiesnews.ycombinator.com/item?id=45817545CSMastermind, Hacker News commenter who evaluates dev tooling for their companyself-described: Decides tooling across their companyTechnical leadershipPublished Nov 5, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Codemaps: Understand Code, Before You Vibe ItID dc-41 | Developer Communitiesnews.ycombinator.com/item?id=45814654bluelightning2k, Hacker News commenter, senior engineerself-described: Senior engineer mixing agentic and regular codingIndividual practitionerPublished Nov 4, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Codemaps: Understand Code, Before You Vibe ItID dc-42 | Developer Communitiesnews.ycombinator.com/item?id=45817024mpalmer, Hacker News commenterself-described: Uses Windsurf at work alongside Claude CodeIndividual practitionerPublished Nov 4, 2025Assessment reviewed Nov 1, 2026Needs a recheck
ServiceNow Says Windsurf Gave Its Engineers a 10% Productivity BoostID sx-02 | Podcasts, Talks & Videosthenewstack.io/servicenow-says-windsurf-gave-its-engineers-a-10-productivity-boost/Pat Casey, CTO and co-founder of ServiceNowEnterprise software company; about 7,000 engineers moved from GitHub CopilotOrganization-widePublished Oct 17, 2025Assessment reviewed Nov 1, 2026Still applicable
[Windsurf Bug] AI keeps getting stuck executing terminal commandsID dc-40 | Developer Communitiesgithub.com/Exafunction/codeium/issues/245joyboy123456, Windsurf user filing a bug reportself-described: Solo, Windows with PowerShellIndividual practitionerPublished Oct 10, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Vibe engineeringID dc-39 | Developer Communitiesnews.ycombinator.com/item?id=45523243tartieret, Hacker News commenter, CTOself-described: Their engineering team, with senior and junior developersTechnical leadershipPublished Oct 9, 2025Assessment reviewed Nov 1, 2026Still applicable
How AI Tools Cut Customer Escalation Time: From Days of Manual Work to MinutesID ld-04 | Engineering Blogs & Reviewsengineering.salesforce.com/how-ai-tools-cut-customer-escalation-time-from-days-of-manual-work-to-minutes/Kristin Foss, VP of Software Engineering for Customer Centric Engineering at Salesforce180-person engineering team resolving customer escalations across Salesforce productsTechnical leadershipPublished Sep 29, 2025Assessment reviewed Nov 1, 2026Still applicable
IDE Comparison with Cursor, Windsurf & Copilot on GPT-5ID ir-07 | Engineering Blogs & Reviewsdev.to/fernandezbaptiste/exploring-cursor-windsurf-and-copilot-with-gpt-5-22blBap (fernandezbaptiste), developerself-described: Solo weekend test on greenfield and brownfield projectsIndividual practitionerPublished Sep 3, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Why does AI feel so different?ID dc-38 | Developer Communitiesnews.ycombinator.com/item?id=44885641scotty79, Hacker News commenterself-described: Solo, a game projectIndividual practitionerPublished Aug 13, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on Hacker News: cancelled Windsurf after the acquisitionID sf-1 | Developer Communitiesnews.ycombinator.com/item?id=44803296cactca, Hacker News commenter and self-described Windsurf subscriberself-described: Solo developerIndividual practitionerPublished Aug 5, 2025Assessment reviewed Nov 1, 2026Current
Windsurf vs. Cursor: When to choose the challengerID ir-06 | Engineering Blogs & Reviewsblog.logrocket.com/windsurf-vs-cursor-when-to-choose-the-challenger/Chizaram Ken, frontend developer writing for LogRocketself-described: Solo, a Next.js to-do app built with CascadeIndividual practitionerPublished Jul 31, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Vibe coding service Replit deleted production database, faked data, told fibsID dc-37 | Developer Communitiesnews.ycombinator.com/item?id=44705100badaigent, Hacker News commenterself-described: Solo, a Next.js appIndividual practitionerPublished Jul 27, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Windsurf In Action: Building AWS InfrastructureID ld-03 | Engineering Blogs & Reviewswww.wwt.com/blog/windsurf-in-action-building-aws-infrastructureSteve Lewis, Lead Software Engineer at World Wide TechnologyTeam building a mobile app backend on AWS with Terraform; WWT also sells Windsurf adoption servicesProfessional teamPublished Jul 23, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Cognition (Devin AI) to Acquire WindsurfID dc-36 | Developer Communitiesnews.ycombinator.com/item?id=44581313justinclift, Hacker News commenterself-described: Their workplace uses Windsurf; the commenter canceled a personal planProfessional teamPublished Jul 16, 2025Assessment reviewed Nov 1, 2026Needs a recheck
How Windsurf Has Revolutionized My Development WorkflowID ld-02 | Engineering Blogs & Reviewswww.wwt.com/blog/how-windsurf-has-revolutionized-my-development-workflowDavid Roff, Software Engineer at World Wide TechnologyFull stack platform team: web, React Native mobile, APIs and databases; WWT also sells Windsurf adoption servicesProfessional teamPublished Jul 14, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleID dc-34 | Developer Communitiesnews.ycombinator.com/item?id=44538404RestlessAPI, Hacker News commenterself-described: Solo; completion-first useIndividual practitionerPublished Jul 12, 2025Assessment reviewed Nov 1, 2026Still applicable
Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleID dc-35 | Developer Communitiesnews.ycombinator.com/item?id=44545308chris_engel, Hacker News commenterself-described: Solo; tried agent workflows in Copilot and WindsurfIndividual practitionerPublished Jul 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleID dc-32 | Developer Communitiesnews.ycombinator.com/item?id=44537303sunaookami, Hacker News commenterself-described: Former Windsurf userIndividual practitionerPublished Jul 11, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI's Windsurf deal is off, and Windsurf's CEO is going to GoogleID dc-33 | Developer Communitiesnews.ycombinator.com/item?id=44537923diegof79, Hacker News commenterself-described: Solo; adopted Windsurf at launchIndividual practitionerPublished Jul 11, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Transcript: 'How Two Engineers Ship Like a Team of 15 With AI Agents'ID ld-08 | Podcasts, Talks & Videosevery.to/podcast/transcript-how-two-engineers-ship-like-a-team-of-15-with-ai-agentsKieran Klaassen, general manager of Cora at EveryTwo-engineer team building Cora, an AI email product at EveryTechnical leadershipPublished Jun 11, 2025Assessment reviewed Nov 1, 2026Still applicable
Lessons in quality engineering from working with Cursor and WindsurfID sx-01 | Engineering Blogs & Reviewswww.ministryoftesting.com/insights/lessons-in-quality-engineering-from-working-with-cursor-and-windsurfMark Winteringham, tester, toolsmith, author and instructorself-described: Solo; migrating his public restful-booker-platform front end to Next.jsIndividual practitionerPublished Jun 10, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: My AI skeptic friends are all nutsID dc-31 | Developer Communitiesnews.ycombinator.com/item?id=44170809osigurdson, Hacker News commenterself-described: Solo, a few months of useIndividual practitionerPublished Jun 3, 2025Assessment reviewed Nov 1, 2026Needs a recheck
TechCrunch: Windsurf says Anthropic is limiting its direct access to Claude AI modelsID sf-3 | Engineering Blogs & Reviewstechcrunch.com/2025/06/03/windsurf-says-anthropic-is-limiting-its-direct-access-to-claude-ai-models/Maxwell Zeff, senior reporter, TechCrunchself-described: Windsurf subscribers using Claude modelsIndividual practitionerPublished Jun 3, 2025Assessment reviewed Nov 1, 2026Current
Comment on: Windsurf SWE-1: Our First Frontier ModelsID dc-30 | Developer Communitiesnews.ycombinator.com/item?id=44027071tianshuo, Hacker News commenterself-described: Paid individual userIndividual practitionerPublished May 19, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-29 | Developer Communitiesnews.ycombinator.com/item?id=44022055robch, Hacker News commenterself-described: Staff-level Android tooling work on a company licenseProfessional teamPublished May 18, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-21 | Developer Communitiesnews.ycombinator.com/item?id=43959991eisfresser, Hacker News commenterself-described: Solo, parallel Cascade sessionsIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-22 | Developer Communitiesnews.ycombinator.com/item?id=43960047pbowyer, Hacker News commenterself-described: Trialled Cursor and Windsurf on the same PHP codebasesIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-23 | Developer Communitiesnews.ycombinator.com/item?id=43960296pembrook, Hacker News commenterself-described: Solo, agentic codingIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-24 | Developer Communitiesnews.ycombinator.com/item?id=43960321powerapple, Hacker News commenterself-described: Solo, a Python projectIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-25 | Developer Communitiesnews.ycombinator.com/item?id=43960966whywhywhywhy, Hacker News commenterself-described: Solo; uses Cursor and WindsurfIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-26 | Developer Communitiesnews.ycombinator.com/item?id=43962096artdigital, Hacker News commenterself-described: Solo; former subscriberIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-27 | Developer Communitiesnews.ycombinator.com/item?id=43962696thesurlydev, Hacker News commenterself-described: Uses the Windsurf plugin inside JetBrains IDEsIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: Cursor or Windsurf?ID dc-28 | Developer Communitiesnews.ycombinator.com/item?id=43962993ksymph, Hacker News commenterself-described: Solo; switches between Cursor and WindsurfIndividual practitionerPublished May 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Ask HN: How much better are AI IDEs vs. copy pasting into chat apps?ID dc-20 | Developer Communitiesnews.ycombinator.com/item?id=43923453raymondgh, Hacker News commenterself-described: Solo, a web app built with WindsurfIndividual practitionerPublished May 8, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI reaches agreement to buy Windsurf for US$3BID dc-18 | Developer Communitiesnews.ycombinator.com/item?id=43912096sagarpatil, Hacker News commenterself-described: Solo; Windsurf customer since launchIndividual practitionerPublished May 7, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI reaches agreement to buy Windsurf for US$3BID dc-19 | Developer Communitiesnews.ycombinator.com/item?id=43912603johntarter, Hacker News commenterself-described: Evaluating Windsurf against GitHub CopilotIndividual practitionerPublished May 7, 2025Assessment reviewed Nov 1, 2026Still applicable
Comment on: OpenAI reaches agreement to buy Windsurf for US$3BID dc-15 | Developer Communitiesnews.ycombinator.com/item?id=43904459zero-g, Hacker News commenterself-described: Tried the editor and the Vim and IntelliJ pluginsIndividual practitionerPublished May 6, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI reaches agreement to buy Windsurf for US$3BID dc-16 | Developer Communitiesnews.ycombinator.com/item?id=43904767__jl__, Hacker News commenterself-described: Uses Cursor and WindsurfIndividual practitionerPublished May 6, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: OpenAI reaches agreement to buy Windsurf for US$3BID dc-17 | Developer Communitiesnews.ycombinator.com/item?id=43904980osigurdson, Hacker News commenterself-described: Solo; writes code the commenter keeps by hand in VimIndividual practitionerPublished May 6, 2025Assessment reviewed Nov 1, 2026Still applicable
Comment on: An update to our pricingID dc-13 | Developer Communitiesnews.ycombinator.com/item?id=43758726tartieret, Hacker News commenter (a CTO, by their own account elsewhere)self-described: Their company, whose developers mostly work in dev containers and WSLTechnical leadershipPublished Apr 22, 2025Assessment reviewed Nov 1, 2026Still applicable
Comment on: An update to our pricingID dc-14 | Developer Communitiesnews.ycombinator.com/item?id=43759232CSMastermind, Hacker News commenter who evaluates dev tooling for their teamsself-described: Evaluates tools for their engineering teamsTechnical leadershipPublished Apr 22, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: An update to our pricingID dc-09 | Developer Communitiesnews.ycombinator.com/item?id=43755492jawns, Hacker News commenterself-described: Individual Windsurf subscriberIndividual practitionerPublished Apr 21, 2025Assessment reviewed Nov 1, 2026Out of dateOut of date, counts toward nothing current
Comment on: An update to our pricingID dc-10 | Developer Communitiesnews.ycombinator.com/item?id=43755988SquareWheel, Hacker News commenterself-described: First week of individual useIndividual practitionerPublished Apr 21, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: An update to our pricingID dc-11 | Developer Communitiesnews.ycombinator.com/item?id=43756387ramesh31, Hacker News commenterself-described: Daily full-time useIndividual practitionerPublished Apr 21, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: An update to our pricingID dc-12 | Developer Communitiesnews.ycombinator.com/item?id=43757246adamgordonbell, Hacker News commenterself-described: Individual, professional workIndividual practitionerPublished Apr 21, 2025Assessment reviewed Nov 1, 2026Out of dateOut of date, counts toward nothing current
Comment on: The Bitter PredictionID dc-08 | Developer Communitiesnews.ycombinator.com/item?id=43674265AndrewKemendo, Hacker News commenter writing as a tech founder planning team sizeself-described: Tech founder sizing engineering teams for future projectsTechnical leadershipPublished Apr 13, 2025Assessment reviewed Nov 1, 2026Needs a recheckListed, no finding position
Matt Wolfe on X: tips after 100+ hours of vibe codingID sx-03 | Developer Communitiesx.com/mreflow/status/1904551509511245921Matt Wolfe, founder of FutureTools and technology video creatorself-described: Solo; more than 100 hours of vibe coding in Cursor and WindsurfIndividual practitionerPublished Mar 25, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Windsurf’s Game-Changing IDE Experience, Part 2, Coding ExamplesID sx-04 | Engineering Blogs & Reviewskeyholesoftware.com/codeium-windsurf-game-changing-ide-experience-part-2/John Hoestje, Software Architect and Developer at Keyhole SoftwareSoftware consultancy; solo experiments building a React site and a Flutter Sudoku gameIndividual practitionerPublished Mar 20, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Cascade stuck in a command execution loop because it cannot execute commandsID dc-07 | Developer Communitiesgithub.com/Exafunction/codeium/issues/153AldrinSalazar, Windsurf user filing a bug reportself-described: Solo, Windows 11Individual practitionerPublished Mar 16, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Vibe coding a nifty Umbraco toolID ir-04 | Engineering Blogs & Reviewsdev.to/skttl/vibe-coding-a-nifty-umbraco-tool-3jicSøren Kottal, Umbraco developerself-described: Solo, a small web tool built in an afternoonIndividual practitionerPublished Mar 16, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: I use Cursor daily - here's how I avoid the garbage partsID dc-06 | Developer Communitiesnews.ycombinator.com/item?id=43342734divan, Hacker News commenterself-described: Daily individual useIndividual practitionerPublished Mar 12, 2025Assessment reviewed Nov 1, 2026Out of dateOut of date, counts toward nothing current
My AI-Powered Workflow for Writing Elixir and Phoenix with WindsurfID ir-03 | Engineering Blogs & Reviewsdev.to/danielbergholz/my-ai-powered-workflow-for-writing-elixir-and-phoenix-with-windsurf-4k8mDaniel Bergholz, developerself-described: Solo side projects in Elixir, Phoenix, React and InertiaIndividual practitionerPublished Mar 10, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: Claude 3.7 Sonnet and Claude CodeID dc-05 | Developer Communitiesnews.ycombinator.com/item?id=43163694bicx, Hacker News commenterself-described: Individual use on new codebasesIndividual practitionerPublished Feb 24, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Remove or increase limit of past conversations in Cascade (Windsurf)ID dc-04 | Developer Communitiesgithub.com/Exafunction/codeium/issues/136nadavoid, Windsurf user filing a feature requestself-described: Solo, Cascade conversationsIndividual practitionerPublished Feb 18, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Comment on: JetBrains Fleet drops support for Kotlin MultiplatformID dc-03 | Developer Communitiesnews.ycombinator.com/item?id=43026200thesurlydev, Hacker News commenterself-described: Uses JetBrains IDEs with Windsurf alongsideIndividual practitionerPublished Feb 12, 2025Assessment reviewed Nov 1, 2026Needs a recheck
The model produced an invalid tool call, trying again.ID sx-05 | Developer Communitiesgithub.com/Exafunction/codeium/issues/125Bwogi, Windsurf subscriber filing a bug reportself-described: Solo; Next.js, TypeScript and Tailwind on macOSIndividual practitionerPublished Jan 30, 2025Assessment reviewed Nov 1, 2026Needs a recheck
Devin Desktop FAQID pd-01 | Product Documentationdocs.devin.ai/desktop/devin-desktop-faqCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Devin Local agentID pd-02 | Product Documentationdocs.devin.ai/desktop/devin-localCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Cascade overviewID pd-03 | Product Documentationdocs.devin.ai/desktop/cascade/cascadeCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Quota-based usageID pd-04 | Product Documentationdocs.devin.ai/desktop/accounts/quotaCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Self-serve plansID pd-05 | Product Documentationdocs.devin.ai/admin/billing/self-serveCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Plans and usageID pd-06 | Product Documentationdocs.devin.ai/desktop/accounts/usageCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Admin controls referenceID pd-07 | Product Documentationdocs.devin.ai/desktop/controlsCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
TerminalID pd-08 | Product Documentationdocs.devin.ai/desktop/terminalCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Memories and rulesID pd-09 | Product Documentationdocs.devin.ai/desktop/cascade/memoriesCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
AGENTS.mdID pd-10 | Product Documentationdocs.devin.ai/desktop/cascade/agents-mdCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Context awarenessID pd-11 | Product Documentationdocs.devin.ai/desktop/context-awareness/windsurf-overviewCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
.devinignoreID pd-12 | Product Documentationdocs.devin.ai/desktop/context-awareness/devin-ignoreCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Agent Command CenterID pd-13 | Product Documentationdocs.devin.ai/desktop/agent-command-centerCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Devin in Devin DesktopID pd-14 | Product Documentationdocs.devin.ai/desktop/devinCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Agent Client ProtocolID pd-15 | Product Documentationdocs.devin.ai/desktop/acpCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Quick ReviewID pd-16 | Product Documentationdocs.devin.ai/desktop/quick-reviewCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Role-based accessID pd-17 | Product Documentationdocs.devin.ai/desktop/accounts/rbac-role-managementCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
SSO and SCIMID pd-18 | Product Documentationdocs.devin.ai/desktop/accounts/sso-scimCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Guide for adminsID pd-19 | Product Documentationdocs.devin.ai/desktop/guide-for-adminsCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only
Get agent outputID pd-20 | Product Documentationdocs.devin.ai/desktop/accounts/api-reference/get-outputCognitionOfficialNot counted (documentation)No date statedDocumentation checked Oct 12, 2026CurrentVerification only