What engineering leaders and their teams report about Seer, with every finding tied to its evidence.
Read the findingsPractitioner sentiment, leaders and teams
+11 Mixed from 29 sources that take a side
How this figure is madeFrom 28 authors in 29 discussions. Small sample. Run a pilot on one production app to measure which failures it surfaces and how long tracing them takes.
Apr 2025 to Oct 2026.
47 sources from practitioners, by month and sentiment.
29 team sources give an opinion: 10 positive, 13 mixed and 6 negative, and most of the mixed ones are bug reports from teams still using Seer. Code review divides between real catches and rejected or blocked reviews. Of the 6 discussing wrong findings, 5 raise them. All 6 discussing platform and setup limits report problems. Teams split on autonomous fixes.
Positive, mixed and negative sources
The score counts them as 9 positive, 13 mixed and 6 negative votes, one per author. See How the numbers work.
See the evidenceBelow: 1 praise theme, 2 concerns, 1 close to agreement and 1 disagreement.
Counted for Engineering leaders and teams, from 31 sources. Change group
Every capability practitioners describe positively, ranked by how many sources say so. Open one to read who said it.
Counted for Engineering leaders and teams.
Positive, 10
Positive
Guilherme Pimenta, Developer at Topsort
Comment on: feat: make an absent user identity something a caller has to stateGitHub pull request review comment
Confirmed and fixed Seer's finding that a deprecation hint would rewrite callers into code that crashes on a blank value.
Positive
Tobias Macey, Engineer at MIT Open Learning
Comment on: feat(dagster): add Sentry error tracking and harden failure alertingGitHub pull request review comment
Fixed four Seer findings on one pull request, including a sensor cursor that would have dropped failures permanently and a fingerprint mismatch that contradicted their own pull request description. The commenter declined one suggested fix as wrong for production alerting and implemented an alternative.
Positive
Ben Atkinson, developer
Probe: check whether Seer AI Code Review runs on this repoGitHub pull request
Opened a probe pull request with deliberate logic and security flaws that passed the linters; Seer's code review ran and posted three findings.
Positive
Tobias Macey, Engineer at MIT Open Learning
Comment on: feat(mit-learn): Fastly edge in front of api.learn.mit.edu, raise webapp HPA capGitHub pull request review comment
Confirmed Seer's finding that an IP restriction would block all non-Fastly traffic before the DNS cutover, and gated it behind the same flag.
Positive
Evie Gauthier, Maintainer of CloudHub-Social Charm
ci: review-gate: require only Seer + Codex (Copilot advisory)GitHub pull request
Made Seer's code review check, with Codex, the required bot review gate on the project's pull requests after Copilot's monthly review credits kept running out.
Mixed
orangemug, Developer at Oak National Academy
Comment on: Switch from npm to pnpmGitHub pull request review comment
First doubted Seer's finding that CI workflows lacked a Node.js setup step after a move to pnpm. A colleague checked the docs a month later, the team agreed it was a real catch, and fixed it.
Mixed
Zamil Majdy, AutoGPT contributor
Comment on: fix(backend/copilot): sync TranscriptBuilder with CLI on mid-stream compactionGitHub pull request review comment
On one pull request the commenter fixed two Seer findings and answered two others as intended behavior.
Positive
Bently, AutoGPT maintainer
Comment on: fix(backend): Extract binary files from ClaudeCodeBlock sandboxGitHub pull request review comment
Fixed a high-severity Seer finding that text files were read into memory without the size check applied to binary files.
Positive
Baptiste Devessier, Contributor to Twenty
Comment on: Support more standard objects in Record Page LayoutGitHub pull request review comment
Accepted Seer's finding that a layout dispatcher change stopped routing the Dashboard object to its renderer, and fixed it before merge.
Positive
etiennejouan, Contributor to Twenty
Comment on: Observability - add new countersGitHub pull request review comment
Acted on Seer's finding that a webhook success counter also counted failed 4xx and 5xx calls, and renamed the metric before merge.
Mixed or negative, 6
Mixed
Charlie, Sentry user
SEER AI Code Review via MCP, SHIFT LEFT!GitHub issue
Their team likes Sentry but finds Seer's pull request reviews slow, and wants the same review in the CLI or IDE before a commit to save CI cycles.
Negative
Timo Juonoja, Developer on the City of Helsinki's Linked Components UI
Comment on: fix: sentry issues fixesGitHub pull request review comment
Rejected two Seer findings rated high and critical severity on the same pull request, a TypeError on an undefined start time and on a missing nested property, as not bugs.
Negative
Nicholas Tindle, Open source maintainer of AutoGPT
Comment on: ci(platform): gate E2E tests behind ci-ready PR labelGitHub pull request review comment
Checked a high-severity Seer finding about a CI trigger against GitHub's documentation, found it wrong, and resolved it as a verified false positive.
Negative
Krzysztof Czerwinski, AutoGPT contributor
Comment on: fix(frontend): AutoPilot notification follow-upsGitHub pull request review comment
Answered a Seer bug report on a notification change by explaining the flagged behavior was intended and did not indicate a bug.
Negative
danilo neves cruz, Engineer at Exactly
Comment on: substreams: initial implementationGitHub pull request review comment
Pushed back on Seer findings as intended design on their team's pull request, where Seer reviewed alongside Cursor, Gemini, and CodeRabbit.
Negative
Seth Herr, Developer on Give Lively's RuboCop rules
Comment on: Update the UniqueIdentifier copGitHub pull request review comment
Rejected a Seer finding marked critical with full confidence, explaining that the matcher it misread leaves out the final quote.
GILD's fact checks, which don't count as sources
Checked Oct 12, 2026.Code Review runs when a pull request is ready and after each subsequent commit. It skips drafts. Use @sentry review to run it on demand. On GitHub, findings produce a neutral status check. Sentry recommends keeping that check optional. Code Review
Open them in the recordEvery concern practitioners raise, most raised first.
Counted for Engineering leaders and teams.
Mixed
stephanefv, Developer at Finverse
Seer Autofix: unable to delete repository from connected repositoriesGitHub issue
Could not disconnect a GitLab repository from Autofix after connecting 12 repositories past a limit of 8; Sentry shipped a rewritten settings backend that fixed it four days later.
Mixed
Matthew Jacobi, GitHub profile lists Oppian and Pitchup
Comment on: Seer GitLab integrationGitHub issue comment
Getting Autofix to run on the GitLab beta took about a developer-day of reading Sentry's source. The engine produced a correct fix and opened a merge request, but webhooks, repository connection, and setup screens broke along the way.
Mixed
Donald Gaeta, IT Software Developer at ARCH Medical
Comment on: Update GitHub Enterprise Integration with the Latest GitHub Integration FeaturesGitHub issue comment
On GitHub Enterprise Cloud their team could not use AI Code Review and other features Sentry advertises for GitHub.com; Sentry replied that Autofix had reached GitHub Enterprise and code review would follow.
Mixed
Øyvind Andreassen, Developer at Elekt
Seer opened a pull request on their team's internal repository although they had not subscribed or asked for one; support traced it to the Show Generative AI Features setting.
Mixed
Yohan Gracia, Developer at Wamiz
Sign commits with Seer AutofixGitHub issue
Their team uses Seer to write Autofix commits but cannot sign them, which they want for security; Sentry support confirmed it is not possible.
Mixed
Tristan Bessoussa, CTO at Wamiz (Nestlé)
Seer "deletage to agent" is not persisting / workingGitHub issue
Switching a project's Autofix handoff from Seer to the Claude Agent did not persist at the organization level; setting it per project worked.
GILD's fact checks, which don't count as sources
Checked Oct 12, 2026.Seer supports GitHub.com and GitLab.com only; self-hosted Git, Bitbucket, and Azure DevOps are not supported, and self-hosted Sentry does not include Seer. Autofix
Open them in the recordCounted for Engineering leaders and teams.
Raised as a concern, 5
Negative
Timo Juonoja, Developer on the City of Helsinki's Linked Components UI
Comment on: fix: sentry issues fixesGitHub pull request review comment
Rejected two Seer findings rated high and critical severity on the same pull request, a TypeError on an undefined start time and on a missing nested property, as not bugs.
Negative
Nicholas Tindle, Open source maintainer of AutoGPT
Comment on: ci(platform): gate E2E tests behind ci-ready PR labelGitHub pull request review comment
Checked a high-severity Seer finding about a CI trigger against GitHub's documentation, found it wrong, and resolved it as a verified false positive.
Negative
Krzysztof Czerwinski, AutoGPT contributor
Comment on: fix(frontend): AutoPilot notification follow-upsGitHub pull request review comment
Answered a Seer bug report on a notification change by explaining the flagged behavior was intended and did not indicate a bug.
Negative
danilo neves cruz, Engineer at Exactly
Comment on: substreams: initial implementationGitHub pull request review comment
Pushed back on Seer findings as intended design on their team's pull request, where Seer reviewed alongside Cursor, Gemini, and CodeRabbit.
Negative
Seth Herr, Developer on Give Lively's RuboCop rules
Comment on: Update the UniqueIdentifier copGitHub pull request review comment
Rejected a Seer finding marked critical with full confidence, explaining that the matcher it misread leaves out the final quote.
Discussed without concern, 1
Mixed
Zamil Majdy, AutoGPT contributor
Comment on: fix(backend/copilot): sync TranscriptBuilder with CLI on mid-stream compactionGitHub pull request review comment
On one pull request the commenter fixed two Seer findings and answered two others as intended behavior.
Statements this group's sources agree on, then the ones close to agreement.
Close to agreement: each needs 7 supporting sources
A person verifies Seer's diagnosis and fixes before merging them.
6 supporting sourcesQuestions the sources split on, with the count on each side.
Evidence remains split
Let it run, 3 sources
Some set Autofix to run automatically or ask for fully autonomous pull requests on high-priority issues.
Stop at the diagnosis, 3 sources
Others cap Seer at root cause or plan, or object to pull requests nobody asked for, because of disclosure and security risk.
The three groups, then every practitioner by the context they work in, and how emphasis changes with scale.
Three groups drawn from the same sources. Choose a group to select whose experience counts. Within that group, each author counts once in the practitioner sentiment, with their latest view. Signals open on Engineering leaders and teams.
31 sources. Team and open source project accounts show Seer's setup limits and costs. They also describe how much authority teams give its automation.
16 sources. Developers report fixes and code review findings from Seer on their own projects.
47 sources. This group combines company and open source team accounts with solo developers. Their platforms and budgets differ.
47 sources from practitioners, by the context they work in. Filled rows count for this group.
Emphasizes Fixes and review findings on their own projects
Emphasizes Review gates and repository-level quirks
Emphasizes Review findings accepted or rejected, platform support, and automation safety
Emphasizes Price and control over fixes
What changes with scale. Individual developers report review catches and describe checking Seer's suggested fixes. Teams report accepted and rejected findings alongside unsupported Git platforms. Their automation decisions also depend on disclosure risks in public repositories and the safety of fixes triggered through a public DSN.
Where this group's sources were published.
Counted for Engineering leaders and teams, from 31 sources. Change group
41 sources for this group, 10 of them documentation
Older accounts remain when they still apply. Accounts tied to legacy per-run pricing or the 2025 Autofix beta carry Needs a recheck.
See sources by originWho each source is, how it knows, whether it still applies and what it counts toward.
Newest first, on its own page. Every count above opens its sources there.
Read all 57 sourcesAbout this read
A running review of what engineering teams report about Seer, from individual builders to engineering leaders. Each finding links to its sources, and each conflict shows both sides.
Counts cover only the sources GILD reviewed.
Choose a group to select whose experience counts. Within that group, each author counts once in the practitioner sentiment, with their latest view. GILD files each source in a group from how the account describes its author. A role or employer marked self-described is what the source states about themselves. Documentation checks facts for every group and never counts toward sentiment.
A source's date and whether it still applies are separate: a recent post can describe an old version, and an old one can still hold.
Flagged for a recheck, of the sources that take a side: leaders and teams 3 of 29, individual developers 1 of 16, all practitioners 4 of 45.
Each row opens its sources for the group you are reading.
| Finding | Engineering leaders and teams | Individual developers | All practitioners | Sources |
|---|---|---|---|---|
| Current readCurrent sentiment | 10 positive, 13 mixed, 6 negative | 10 positive, 5 mixed, 1 negative | 20 positive, 18 mixed, 7 negative | Sources |
| PraiseCode Review Catches | 10 of 16 positive | 5 of 7 positive | 15 of 23 positive | Sources |
| PraiseRoot Cause Analysis | 1 of 1 positive (under the minimum) | 1 of 1 positive (under the minimum) | 2 of 2 positive (under the minimum) | Sources |
| PraiseFix Quality | 2 of 3 positive (under the minimum) | 2 of 2 positive (under the minimum) | 4 of 5 positive | Sources |
| PraiseTime to an Answer | 1 of 2 positive (under the minimum) | 1 of 1 positive (under the minimum) | 2 of 3 positive (under the minimum) | Sources |
| ConcernPlatform & Setup Limits | 6 of 6 raise it | 3 of 3 raise it (under the minimum) | 9 of 9 raise it | Sources |
| ConcernPer-Contributor Price | 1 of 2 raise it (under the minimum) | 0 of 1 raise it (under the minimum) | 1 of 3 raise it (under the minimum) | Sources |
| ConcernData & Automation Safety | 3 of 3 raise it (under the minimum) | 1 of 2 raise it (under the minimum) | 4 of 5 raise it | Sources |
| ConcernWrong Findings | 5 of 6 raise it | 1 of 1 raise it (under the minimum) | 6 of 7 raise it | Sources |
| ConcernUnrequested Output | 1 of 2 raise it (under the minimum) | 0 of 0 raise it (under the minimum) | 1 of 2 raise it (under the minimum) | Sources |
| ConsensusSeer works with a coding agent | 2 supporting (under the minimum) | 1 supporting (under the minimum) | 3 supporting (under the minimum) | Sources |
| ConsensusA person checks Seer's output | 6 supporting (below threshold) | 3 supporting (under the minimum) | 9 supporting (below threshold) | Sources |
| ConsensusTriage volume is the reason to automate | 2 supporting (under the minimum) | 1 supporting (under the minimum) | 3 supporting (under the minimum) | Sources |
| DisagreementShould diagnosis run in Seer, or in a coding agent the team already uses? | Seer inside Sentry 4, Their own agent 0 (under the minimum) | Seer inside Sentry 1, Their own agent 1 (under the minimum) | Seer inside Sentry 5, Their own agent 1 (under the minimum) | Sources |
| DisagreementHow far should Seer act without a person starting it? | Let it run 3, Stop at the diagnosis 3 | Let it run 1, Stop at the diagnosis 0 (under the minimum) | Let it run 4, Stop at the diagnosis 3 | Sources |
| Current readEvidence mix | 41 sources | 26 sources | 57 sources | Sources |
GILD's checks of claims against official documentation, kept apart from the source counts.
