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.
Source
59 sources in All practitioners, with product documentation Showing all 59 sources
No opinion stated
Anthropic S-1Hacker News comment
The author expects enterprises to keep buying from cloud platforms, such as Foundry with zero data retention, before they would run their own GPU clusters.
Mixed
The author calls the Foundry platform really good but feels every call through it is about twice as slow.
Mixed
Stephen W. Thomas on X: "Microsoft Foundry AI model deployment options confused me!"Post on X
Foundry's model deployment options confused the author enough to record a walkthrough of Global, Data Zone, and Standard deployments and how region affects which models are available.
Not counted
Its author is a Microsoft MVP who publishes Azure training content. Kept in the record; it takes part in no figure.
Positive
IBM CEO says open weight models are now good and cheap on-premHacker News comment
The author argues that companies should rent inference and skip buying GPUs, because Foundry and Amazon Bedrock already offer many open and closed models with zero data retention.
Mixed
Bedrock vs Vertex AI vs Foundry: Your Cloud Commit DecidesNewsletter analysis, THE DAILY BRIEF
The author rates Foundry the strongest of the three platforms for GPT models, with the most flexible provisioned throughput, and the weakest for Claude: no Claude provisioned throughput, no EU data zone, and Anthropic as the data processor. At list prices the platforms differ by at most about US$300 a month on a fixed Claude workload, so existing committed spend should decide.
Counts toward
Mixed
The author says Foundry can feel at first like a long list of models, agents, tools, evaluations, guardrails, and governance features.
Positive
DeepInfer: High-performance inference framework for DeepSeek modelsHacker News comment
The author can deploy open-weight models in Foundry with one click, already approved by the company's compliance and procurement teams. Data contracts are why the author expects cloud platforms to win enterprise model business.
Positive
OpenCode 2.0Hacker News comment
The author's organization of more than 100 developers hosts its own models in Foundry and points Claude Code at them. Developers need no API keys because they sign in with az login.
Negative
[Azure.AI.AgentServer.Responses] response.completed is streamed before the conversation lock is released, so the next turn gets 409 conversation_lockedGitHub issue, Azure/azure-sdk-for-net
A Foundry hosted agent streams its completion event before it releases the conversation, which stays locked for a further 0.5 to 3.7 seconds. An immediate follow-up turn fails with 409 conversation_locked.
No opinion stated
Why isn't the industry freaking out about DeepSeek 4.1 Flash?Hacker News comment
The author's large enterprise bans Chinese models outright and does not yet treat a Chinese model hosted in Foundry any differently.
Counts toward
Positive
Using Opus 5.5 to discover a new eyewitness record of the dodoHacker News comment
The author says Foundry has every frontier model except Google's, so a Foundry subscription plus a Google one covers a workflow that compares several models on the same input.
Mixed
GPT 5.6 Sol Still Fails on Basic Code ReviewHacker News comment
Checking the Foundry catalog, the author found that the separate reasoning variants of a model seemed to be gone, leaving what may be one model.
Negative
azure-ai-projects: gpt-6-luna returns 500 on project endpoint, works on resource endpointGitHub issue, Azure/azure-sdk-for-python
A newly released model returned HTTP 500 errors on the Foundry project endpoint while working on the resource endpoint, and the failure reproduced with raw HTTP. Other models in the same project worked on both.
Negative
.NET: [FoundryAgent][gpt-6-luna]: [Bug]: Invalid parameter: 'text.format' of type 'json_schema' is not supported with model version gpt-6-luna-2026-09-22GitHub issue, microsoft/agent-framework
A Foundry agent that worked on one model failed with a structured output error after switching to a newer one, and the reasoning effort setting could not be removed in the Foundry agent interface.
Positive
Grok 4.7Hacker News comment
The author deployed DeepSeek V4.1 Flash on the team's Foundry resource, ran it through Pi, and liked its tone.
Negative
.NET: [Bug]: Hosted-agent runtime never activates OTLP export to a third-party destination, despite docs describing it as automaticGitHub issue, microsoft/agent-framework
A .NET Foundry hosted agent never sent OpenTelemetry data to a third-party destination through the documented environment variables, which work outside the hosted runtime.
Not counted
Its author is a Microsoft MVP, a recognition program Microsoft runs. Kept in the record; it takes part in no figure.
No opinion stated
DeepSeek v4.1 FlashHacker News comment
The author explains that changing or removing Foundry's safety guardrails and content filters is a Limited Access feature that requires a request form.
Negative
DeepSeek v4.1 FlashHacker News comment
The author finds Foundry slow to add new models and missing many open-weight ones.
Positive
Navier-Stokes - Tristan Buckmaster [pdf]Hacker News comment
For teams that do not trust the model vendors with their data, the author points to Foundry or Amazon Bedrock, which earn money from renting infrastructure and so have a stronger reason not to train on customer data.
Negative
Baptiste GHO on X: "background mode is disabled for every model we've tried in Azure AI Foundry"Post on X
Background mode was disabled for every model the company tried in Foundry, which blocked a new course generation feature in production.
Negative
GPT-6 Astra on OpenRouterHacker News comment
The author says zero data retention is the main reason customers pick Foundry over OpenAI's own API, yet the model list makes it hard to see which models carry it. There is no simple way to block models without it, so each customer vets and approves models one by one.
No opinion stated
GPT-6 AstraHacker News comment
The author notes that the latest OpenAI models have kept arriving in Foundry.
Positive
OpenRouter is joining StripeHacker News comment
The author says corporate compliance teams reject third-party model routers and recommends coding enterprise apps with fallback loops across cloud platforms such as Foundry and Amazon Bedrock, which offer private endpoints and legal indemnification.
No opinion stated
Kimi K2.7 Code is generally available in GitHub CopilotHacker News comment
The author notes that the DeepSeek V4 models were already in Foundry.
No opinion stated
Kimi K2.7 Code is generally available in GitHub CopilotHacker News comment
Quoting GitHub's documentation, the author points out that these models run on US-based Foundry infrastructure and that prompts and responses are not sent to the original model developers.
Mixed
Kimi K2.7 Code is generally available in GitHub CopilotHacker News comment
Adding a Foundry model as a GitHub Copilot custom model failed for the author, and the path is poorly documented and unstable. VS Code's bring-your-own-key route worked, so the author now uses DeepSeek V4 Flash with Copilot.
Positive
I built a WhatsApp clone of myself; a friend agreed on a password with itHacker News comment
The author built a WhatsApp bot that answers as the author, with Claude Sonnet 4.6 through Foundry generating each reply, in about two hours of work. Runtime cost is about US$0.005 per reply.
Negative
Azure AI Foundry Agent published to Microsoft Teams prompts users for Foundry login instead of using Teams SSOMicrosoft Q&A thread
The author published a Foundry agent to Teams for users in his organization's tenant. On first use each user gets a sign-in card that sends them to Foundry's own sign-in service before the agent answers, where Copilot Studio agents start without one. A Microsoft moderator replied on Jun 18, 2026 that this first sign-in is expected and that no documented setting removes it.
Positive
Cybersecurity researchers aren't happy about the guardrails on Anthropic's FableHacker News comment
The author deploys Claude models on Azure because clients already trust their Microsoft contracts, and says the Anthropic models make the products much better. The author asks how data terms for newer frontier models affect that trust.
Mixed
Decisions that eroded trust in Azure - by a former Azure Core engineerHacker News comment
The author's team has run critical services on Azure for ten years without a major incident, but calls the AI Foundry side broken at the moment, as are other vendors' AI services.
Negative
Runtime validation is still fucked in AI coding agentsHacker News comment
Two days of setting up AI end-to-end tests with vision calls in Foundry ran up a bill of more than US$100, which surprised the author's manager.
Mixed
Create Preview web experience from new AI Foundry is incompleteGitHub issue, microsoft-foundry/foundry-agent-webapp
The settings file the new Foundry portal tells users to create for an agent's web app was missing required variables, so the steps did not work as written.
Not counted
Its author works at Zure, a Microsoft partner consultancy. Kept in the record; it takes part in no figure.
Negative
[Bug] 400 Error when chatting with an Agent published to Teams/Web using new foundry portalGitHub issue, microsoft-foundry/new-foundry-portal
Two agents that worked in the playground and published without error returned 400 Bad Request in Teams and web chat, with no traces to debug.
Negative
Getting a Gemini API key is an exercise in frustrationHacker News comment
The author sees Foundry serving return HTTP 500 errors for stretches of time and advises production teams to pick models available from several clouds and fail over between them.
Negative
Getting a Gemini API key is an exercise in frustrationHacker News comment
The author found the setup confusing: a choice between Azure OpenAI and Microsoft Foundry resources, a separate portal with an old and a new Foundry experience, Foundry and Foundry Project resource types, and a single API key unless Entra ID is set up.
Mixed
Microsoft has a problem: lack of demand for its AI productsHacker News comment
The author calls Azure's AI services decent and access to OpenAI models through Foundry a real differentiator, but says rate limits are a problem. The author worries the Foundry rebranding signals marketing taking over the services.
Negative
Google AntigravityHacker News comment
The author says enterprise work is forced onto cloud platforms such as Foundry, which do the same job as the vendors' own APIs with a much worse developer experience, and calls their extra prompt management and evaluation features poor.
Negative
Microsoft's template for deploying Foundry on a private network failed both on a work subscription with an existing virtual network and on a clean personal subscription.
Negative
The LLM Lobotomy?Hacker News comment
Foundry still lacked tool calling for Llama models more than six months after Microsoft said it was working on it, and small models were consistently slow. Provisioned throughput would fix speed but started at about $10k a month for GPT models and about $72k for DeepSeek, against a few dollars a month of current use.
Counts toward, pending a recheck
Negative
azure-ai-agents connected agent with MCP will raise for approval and eventually crashGitHub issue, Azure/azure-sdk-for-python
When a coordinator agent invoked a connected agent that used an MCP tool, the run hung in progress and then failed with a premature end of the response.
Counts toward, pending a recheck
Negative
azure-ai-agents: MCP usage errors with Azure AI FoundryGitHub issue, Azure/azure-sdk-for-python
Foundry agents frequently raised errors when using MCP tools and repeatedly failed to call them. The issue drew 47 comments.
Counts toward, pending a recheck
Negative
Azure AI Foundry Agent Service Deep Research Tool ErrorGitHub issue, Azure/azure-sdk-for-python
With access to o3-deep-research granted, Microsoft's own example code for the Deep Research tool still failed with an error that the tool was not supported.
Negative
Tool call of AzureAISearchTool doesn't respect tool configuration such as filter and top_k in package azure-ai-agentsGitHub issue, Azure/azure-sdk-for-python
The agent's Azure AI Search tool ignored the filter and top_k settings configured on it.
Counts toward, pending a recheck
Negative
400 reasons to not use Microsoft AzureHacker News comment
Prompt flow went down for a day with no status or information, and creating a new pay-as-you-go deployment meant switching back to the old portal to find the button.
Negative
DeepSeek R1 Is Now Available on Azure AI Foundry and GitHubHacker News comment
DeepSeek R1 on Foundry took 80 seconds to answer a question GPT-4o answered in 7, under rate limits and a 4,000-token input cap at launch. The author also found errors in the inference SDK's documentation and choppy streaming from Mistral Large.
Counts toward, pending a recheck
Pages from Microsoft, used to check facts for every group.
States that Foundry groups agents, models, and tools under one management grouping with tracing, monitoring, and evaluations, governed with Microsoft Entra identity, role-based access control, content filters, network isolation, and Azure Policy. Lists access to more than 10,000 models from Microsoft, OpenAI, Anthropic, Meta, and others, and tells readers to choose a model from the Foundry Models catalog.
States that Foundry supports sign-in with Microsoft Entra ID and API keys, and recommends Entra ID for production because it enables conditional access, multifactor authentication, managed identities, and least-privilege role-based access control. API keys lack per-user traceability.
Describes how Foundry Agent Service integrates with tools through one MCP-compatible endpoint, with credentials handled through Microsoft Entra ID and OAuth identity passthrough. Lists MCP servers, Azure AI Search, OpenAPI, and Fabric IQ as tools agents use through a toolbox or directly, and SharePoint and the Microsoft Fabric data agent as direct integrations.
Describes a model catalog of over 10,000 models, with about 50 added each month, from vendors including Azure OpenAI, Anthropic, Mistral, Meta, Cohere, and NVIDIA, and invites teams to bring your own model or use a hosted one. Serverless deployments are standard pay-per-token or provisioned reserved capacity, with global, data zone, or regional processing.
Describes model router, deployed like any other Foundry model, which routes each prompt to a suitable underlying model and does not store prompts. Balanced is the default routing mode, with Cost and Quality as alternatives, on Global Standard and Data Zone Standard deployments.
Lists pay-as-you-go prices per million tokens in US dollars for DeepSeek models in East US 2, for example DeepSeek-V4 Flash Global at $0.19 input and $0.51 output, and DeepSeek R1 Global at $1.35 input and $5.40 output. Provisioned throughput for DeepSeek R1 and V3 starts at 100 PTUs, at $1 per PTU hour or $260 per PTU month reserved.
Documents spillover, which sends overage requests from a fully used provisioned deployment to a standard deployment in the same Foundry resource. It requires an active pay-as-you-go deployment of the same model and version.
Shows how to send a Foundry resource's Audit Logs, Request and Response Logs, and usage logs through diagnostic settings to Azure Storage, which retains them for policy audit, or to Log Analytics, where Azure Monitor records can be queried.
Lists Azure OpenAI and Foundry Tools services such as Azure AI Search as in scope for FedRAMP High and DoD IL2 in Azure public regions. In Azure Government the Microsoft Foundry portal is in scope for FedRAMP High and DoD IL2.
States that Azure is in scope for Microsoft's SOC 2 Type 2 attestation report, with service-level detail in the Microsoft Azure Compliance Offerings document.
Lists Azure, Azure Government, and Azure Germany among the Microsoft cloud services in scope for its ISO/IEC 27001 certification.
States that prompts, completions, embeddings, and training data are not used to train generative AI foundation models without the customer's permission, and that models are stateless. Global deployments may process prompts in any geography where the model is deployed, and Data Zone deployments only inside the data zone.
States that quotas apply per Azure subscription and region, not per tenant, with up to 100 Foundry resources per region per subscription, 250 projects per resource, and 32 deployments per resource. Tokens-per-minute increases go through a request form and are evaluated one by one.
States that the platform moved from Azure AI Studio to Azure AI Foundry to Microsoft Foundry on the same Microsoft.CognitiveServices/accounts resource type. The azure-ai-inference package and the Assistants API retire on August 26, 2026, with agents moving to the Responses API, which is not available in every Azure region.
Source index
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.
| Source | Original URL | Author and context | Who | Published | Review date | Applies | Findings |
|---|---|---|---|---|---|---|---|
| Anthropic S-1ID s-13 | Developer Communities | news.ycombinator.com/ | 827a, developerself-described: Discussion of whether enterprises will run their own model hardware | Individual builder | Published Oct 27, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| VanillaCache on X: "for azure foundry, the platforms really good, but it's a pity I feel like any calls through it are just 2X slower"ID s-41 | Developer Communities | x.com/ | VanillaCache, developerself-described: Calls models through Foundry and other hosts | Individual builder | Published Oct 27, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Stephen W. Thomas on X: "Microsoft Foundry AI model deployment options confused me!"ID s-42 | Developer Communities | x.com/ | Stephen W. Thomas, Azure integration architect and Microsoft MVPself-described: Walkthrough of Foundry deployment types and regional model availability | Individual builder | Published Oct 27, 2026 | Assessment reviewed Oct 28, 2026 | Current | Not counted. Its author is a Microsoft MVP who publishes Azure training content. |
| IBM CEO says open weight models are now good and cheap on-premID s-11 | Developer Communities | news.ycombinator.com/ | fg137, developerself-described: Discussion of whether to rent model capacity or buy GPUs | Individual builder | Published Oct 26, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Bedrock vs Vertex AI vs Foundry: Your Cloud Commit DecidesID s-44 | Blogs & Newsletters | www.beri.net/ | Rajesh Beri, enterprise AI practitioner and author of THE DAILY BRIEFself-described: Platform comparison for enterprise buyers choosing where to run Claude, GPT, and Gemini | Technical leadership | Published Oct 25, 2026 | Assessment reviewed Oct 28, 2026 | Current | Overall signalModel Catalog & Choice: Mixed or negativeContracts, Data Terms & Identity: PositiveCost: PositiveMissing Models & Features: Raised as a concernData Terms & Model Governance: Raised as a concernQuotas, Capacity & Cost: Raised as a concernChosen for the contract and data termsDoes Foundry's catalog have the models a team needs?: Gaps and delays |
| AiswaryaVenkit1 on X: "At first glance, it can feel like a long list of models, agents, tools, evaluations, guardrails, and governance features."ID s-43 | Developer Communities | x.com/ | AiswaryaVenkit1, developerself-described: Thread explaining what Microsoft Foundry includes | Individual builder | Published Oct 23, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| DeepInfer: High-performance inference framework for DeepSeek modelsID s-02 | Developer Communities | news.ycombinator.com/ | kasey_junk, developerself-described: Company whose compliance and procurement teams approve the models developers may use | Individual builder | Published Oct 20, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| OpenCode 2.0ID s-01 | Developer Communities | news.ycombinator.com/ | cdnsteve, developerself-described: Engineering organization of more than 100 developers running Claude Code against its own Foundry deployments | Organization-wide | Published Oct 15, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
[Azure.AI.AgentServer.Responses] response.completed is streamed before the conversation lock is released, so the next turn gets 409 conversation_lockedID s-07 | Developer Communities | github.com/ | guramrit-dhillon, developerself-described: Hosted agent on Foundry Agent Service built with Microsoft Agent Framework on .NET 9 | Individual builder | Published Oct 15, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Why isn't the industry freaking out about DeepSeek 4.1 Flash?ID s-20 | Developer Communities | news.ycombinator.com/ | browningstreet, developer at a large enterpriseself-described: Large enterprise with a policy on Chinese models | Organization-wide | Published Oct 8, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Using Opus 5.5 to discover a new eyewitness record of the dodoID s-05 | Developer Communities | news.ycombinator.com/ | jiggawatts, consultantself-described: Multi-model OCR workflow that runs the same prompt through several models | Individual builder | Published Oct 2, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| GPT 5.6 Sol Still Fails on Basic Code ReviewID s-21 | Developer Communities | news.ycombinator.com/ | chris_money202, developerself-described: Compares model behavior across coding tools | Individual builder | Published Sep 28, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| azure-ai-projects: gpt-6-luna returns 500 on project endpoint, works on resource endpointID s-06 | Developer Communities | github.com/ | Konnor (konnor-willison-deepsee), developerself-described: Python application calling models through a Foundry project in Sweden Central | Individual builder | Published Sep 24, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
.NET: [FoundryAgent][gpt-6-luna]: [Bug]: Invalid parameter: 'text.format' of type 'json_schema' is not supported with model version gpt-6-luna-2026-09-22ID s-38 | Developer Communities | github.com/ | lucianteodorescu-milestone, developerself-described: Foundry agent switched to a newer model | Individual builder | Published Sep 24, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Grok 4.7ID s-03 | Developer Communities | news.ycombinator.com/ | taspeotis, developerself-described: Team running non-OpenAI models from its own Foundry resource in the Pi coding harness | Individual builder | Published Sep 21, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| .NET: [Bug]: Hosted-agent runtime never activates OTLP export to a third-party destination, despite docs describing it as automaticID s-37 | Developer Communities | github.com/ | Cristopher Coronado, Senior AI Engineer and Microsoft MVPself-described: .NET hosted agent on Foundry sending telemetry to a third-party destination | Individual builder | Published Sep 15, 2026 | Assessment reviewed Oct 28, 2026 | Current | Not counted. Its author is a Microsoft MVP, a recognition program Microsoft runs. |
| DeepSeek v4.1 FlashID s-08 | Developer Communities | news.ycombinator.com/ | HanClinto, developerself-described: Discussion of how to reach models with fewer safety filters | Individual builder | Published Sep 10, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| DeepSeek v4.1 FlashID s-09 | Developer Communities | news.ycombinator.com/ | cdnsteve, developerself-described: Engineering organization that hosts its coding models in Foundry | Individual builder | Published Sep 10, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Navier-Stokes - Tristan Buckmaster [pdf]ID s-10 | Developer Communities | news.ycombinator.com/ | edot, developerself-described: Discussion of whether model vendors train on customer data | Individual builder | Published Sep 9, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Baptiste GHO on X: "background mode is disabled for every model we've tried in Azure AI Foundry"ID s-40 | Developer Communities | x.com/ | Baptiste GHO, CTO, Complement Learningself-described: Course generation feature in production at Complement Learning | Technical leadership | Published Sep 9, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| GPT-6 Astra on OpenRouterID s-04 | Developer Communities | news.ycombinator.com/ | jiggawatts, consultantself-described: Advises customers who choose between Foundry and OpenAI's own API | Individual builder | Published Sep 5, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| GPT-6 AstraID s-14 | Developer Communities | news.ycombinator.com/ | jiocrag, developerself-described: Discussion of where new OpenAI models are available | Individual builder | Published Sep 3, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| OpenRouter is joining StripeID s-12 | Developer Communities | news.ycombinator.com/ | johnbarron, developerself-described: Corporate AI work weighing model routers against cloud platforms | Individual builder | Published Aug 20, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Kimi K2.7 Code is generally available in GitHub CopilotID s-15 | Developer Communities | news.ycombinator.com/ | pkaye, developerself-described: Discussion of which models reach Microsoft's platforms | Individual builder | Published Jul 2, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Kimi K2.7 Code is generally available in GitHub CopilotID s-16 | Developer Communities | news.ycombinator.com/ | rombert, developerself-described: Discussion of where Chinese open-weight models run in GitHub Copilot | Individual builder | Published Jul 2, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Kimi K2.7 Code is generally available in GitHub CopilotID s-17 | Developer Communities | news.ycombinator.com/ | mvATM99, developerself-described: Connects a Foundry model to GitHub Copilot in VS Code | Individual builder | Published Jul 2, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| I built a WhatsApp clone of myself; a friend agreed on a password with itID s-18 | Developer Communities | news.ycombinator.com/ | dandinu, developerself-described: Personal WhatsApp reply bot with retrieval over the author's past messages | Individual builder | Published Jun 19, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Azure AI Foundry Agent published to Microsoft Teams prompts users for Foundry login instead of using Teams SSOID s-45 | Developer Communities | learn.microsoft.com/ | Jino George, who published a Foundry agent to Microsoft Teams for users in his organization's Microsoft 365 tenantself-described: A Foundry agent published straight to Microsoft Teams for users in the same Microsoft 365 tenant | Product or platform team | Published Jun 15, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Cybersecurity researchers aren't happy about the guardrails on Anthropic's FableID s-19 | Developer Communities | news.ycombinator.com/ | consumer451, developer building products for clientsself-described: Client products built on Claude models deployed in Azure | Individual builder | Published Jun 11, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Decisions that eroded trust in Azure - by a former Azure Core engineerID s-29 | Developer Communities | news.ycombinator.com/ | mrsmrtss, developerself-described: Team that has run critical services on Azure for ten years | Small product team | Published Apr 3, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Runtime validation is still fucked in AI coding agentsID s-30 | Developer Communities | news.ycombinator.com/ | sebringj, developerself-described: Team trying AI end-to-end testing at the request of the author's manager | Small product team | Published Feb 10, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Create Preview web experience from new AI Foundry is incompleteID s-39 | Developer Communities | github.com/ | Jan de Vries, engineer, Zureself-described: Publishing a Foundry agent as a web app from the new portal | Product or platform team | Published Jan 28, 2026 | Assessment reviewed Oct 28, 2026 | Current | Not counted. Its author works at Zure, a Microsoft partner consultancy. |
| [Bug] 400 Error when chatting with an Agent published to Teams/Web using new foundry portalID s-35 | Developer Communities | github.com/ | Alexander Danson, Product Manager, Marshall Groupself-described: Agents built in the new Foundry portal and published to Microsoft Teams | Product or platform team | Published Jan 9, 2026 | Assessment reviewed Oct 28, 2026 | Current | |
| Getting a Gemini API key is an exercise in frustrationID s-22 | Developer Communities | news.ycombinator.com/ | hobofan, developerself-described: Production AI features that need high uptime | Individual builder | Published Dec 11, 2025 | Assessment reviewed Oct 28, 2026 | Still applicable | |
| Getting a Gemini API key is an exercise in frustrationID s-23 | Developer Communities | news.ycombinator.com/ | arielcostas, developerself-described: Project that needed Azure speech voices and Foundry resources | Individual builder | Published Dec 10, 2025 | Assessment reviewed Oct 28, 2026 | Still applicable | |
| Microsoft has a problem: lack of demand for its AI productsID s-24 | Developer Communities | news.ycombinator.com/ | zubiaur, developerself-described: Builds on Azure AI services, including AI Search and Document Intelligence | Individual builder | Published Dec 8, 2025 | Assessment reviewed Oct 28, 2026 | Still applicable | |
| Google AntigravityID s-25 | Developer Communities | news.ycombinator.com/ | koakuma-chan, developerself-described: Enterprise work that must reach models through a cloud platform | Individual builder | Published Nov 18, 2025 | Assessment reviewed Oct 28, 2026 | Still applicable | |
| Cannot deploy on Private Network: Failed to create Aml RP virtual workspace due to System.Exception: Failed async operationID s-36 | Developer Communities | github.com/ | Daryl Graves, developerself-described: Deploying the standard agent setup on a private network from Microsoft's template | Individual builder | Published Sep 23, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| The LLM Lobotomy?ID s-26 | Developer Communities | news.ycombinator.com/ | Julian Goldsmith (juliangoldsmith), developerself-described: Ongoing project with low request volume on Foundry | Individual builder | Published Sep 20, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| azure-ai-agents connected agent with MCP will raise for approval and eventually crashID s-33 | Developer Communities | github.com/ | Ilya Rakhlin (irakhlin), developerself-described: Multi-agent setup with a coordinator agent and a connected agent, each with an MCP tool | Individual builder | Published Aug 17, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| azure-ai-agents: MCP usage errors with Azure AI FoundryID s-31 | Developer Communities | github.com/ | reesmonty (jacobreesmontgomery), developerself-described: Foundry agents calling remote MCP servers such as GitMCP and Tavily | Individual builder | Published Aug 8, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| Azure AI Foundry Agent Service Deep Research Tool ErrorID s-32 | Developer Communities | github.com/ | Sashank Neelamraju (sashanknjs), developerself-described: Trying the Deep Research tool in Foundry Agent Service | Individual builder | Published Jul 8, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
Tool call of AzureAISearchTool doesn't respect tool configuration such as filter and top_k in package azure-ai-agentsID s-34 | Developer Communities | github.com/ | Vitor Arrais, engineer, N26self-described: Retrieval agent over an Azure AI Search index | Product or platform team | Published Jul 4, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| 400 reasons to not use Microsoft AzureID s-27 | Developer Communities | news.ycombinator.com/ | Oras, developerself-described: A few months building on Azure AI Foundry and AI Search | Individual builder | Published Mar 1, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| DeepSeek R1 Is Now Available on Azure AI Foundry and GitHubID s-28 | Developer Communities | news.ycombinator.com/ | vemonet, developerself-described: Testing DeepSeek R1 and Mistral Large through serverless APIs on a paid subscription | Individual builder | Published Jan 30, 2025 | Assessment reviewed Oct 28, 2026 | Needs a recheck | |
| What is Microsoft Foundry?ID pd-01 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Authentication and authorization in Microsoft FoundryID pd-02 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| What is Toolbox in Microsoft Foundry?ID pd-03 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Microsoft Foundry Models overviewID pd-04 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Model router for Microsoft Foundry conceptsID pd-05 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Foundry Models PricingID pd-06 | Product Documentation | azure.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Manage traffic with spillover for provisioned deploymentsID pd-07 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Enable diagnostic logging for Microsoft FoundryID pd-08 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Azure and other Microsoft cloud services compliance scopeID pd-09 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| System and Organization Controls (SOC) 2 Type 2ID pd-10 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| ISO/IEC 27001:2013 Information Security Management StandardsID pd-11 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Data, privacy, and security for Foundry Models sold by Azure in Microsoft FoundryID pd-12 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Microsoft Foundry Models quotas and limitsID pd-13 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
| Migrate from the Foundry (classic) portalID pd-14 | Product Documentation | learn.microsoft.com/ | MicrosoftOfficial | Not counted (documentation) | No date stated | Documentation checked Oct 28, 2026 | Current | Verification only |
