← Feed

Posts

• Post #4480179

  Just discovered Hillel Wayne's Crossover Project, where he interviews a bunch of "trad" engineers who became SWEs, to decide if SWE is really "engineering". I know we all have gut reactions, but read it! https://www.hillelwayne.com/tags/crossover-project/

• Post #4280189

  My Posette talk is now on YT- "An MCP for your Postgres DB" https://youtube.com/watch?v=3_JPHuXgDyQ I build a Python FastMCP server, starting with free-form SQL tools, then read-only tools, then typed query-specific tools. Slides: https://pamelafox.github.io/mcp-for-postgres-db-demo/ Code: https://github.com/pamelafox/mcp-for-postgres-db-demo

• Post #4280188

  I love my cute lil iPhone 13 mini that fits in all my pockets, but its battery is on its last legs. What's the dantiest modern iPhone I can get?

• Post #4280187

  My colleague and I are running a new livestream series: Microsoft IQ Deep Dive with Python We'll cover: * Foundry IQ (Azure AI Search): create multi-source knowledge bases with hybrid search and re-ranking * Fabric IQ: query ontologies from MS Fabric * Work IQ: query M365 data like teams, chats, calendar events, etc 🗓️ July 28–30, 10 AM PT 👉🏼 Register: https://aka.ms/IQDeepDivePython/series

• Post #4280186

  Super interesting paper: "Prompt Injection as Role Confusion" https://role-confusion.github.io/ They trained a linear probe to predict roles (user/tool/assistant), and then experimented with injecting role tags and just changing the writing style to mimic a role (like <think>ing).

• Post #4280185

  my 7-yr-old wants a karaoke machine for her bday! how do karaoke machines work in the modern age? notably, we have no TV, only a projector with chromecast. any recommendations?

• Post #4280184

  My current guilty pleasure is watching “Love is Blind: Argentina” - but I figure it’s educational, since it’s all in spanish! Argentinian spanish is a fascinating dialect, it has a lot of sound changes: https://en.wikipedia.org/wiki/Rioplatense_Spanish

• Post #4280183

  As a life-long Photoshop addict, I love that the web is now powerful enough for basic photo editing operations. I've been using http://photopea.com, as it's free and mimics the Photoshop UI and shortcuts.

• Post #4280182

  Great intro to "product quantization", a vector compression technique used by ColBERT. https://isaacflath.com/writing/QuantizationFundamentalsForMultiVectorRetrieval

• Post #4280181

  Great intro to git worktrees from @nzakas: https://humanwhocodes.com/blog/2026/07/introduction-git-worktrees/ The GitHub Copilot app uses worktrees by default, and it reallly helps to understand them, since a worktree will *not* have files from the main checkout like a gitignored .env.

• Post #4280180

  "Do automated evals work?" https://parlance-labs.com/blog/posts/auto-evals/ Hamel and Antaripa use multiple LLM-powered tools to identify failures in app traces and compare results to labels from a domain expert. LLMs caught some errors that humans missed, but missed big errors that humans flagged. Read the post for their recommendations.

• Post #4280179

  Released new version of axe-core-playaxe-playwright-python for automated a11y testing of websites: https://pypi.org/project/axe-playwright-python/ Thank you to the community for their PRs keeping it up to date with latest axe-core accessibility checks.

• Post #4280178

  The new MCP spec is officially out! https://blog.modelcontextprotocol.io/posts/2026-07-28/ To celebrate, my colleague and I are hosting a 5-hour livestream on September 9 on the VS Code YouTube channel. Join MCP maintainers and builders for talks about MCP Authorization, Tasks, Apps, and more. Register @ https://developer.microsoft.com/en-us/reactor/events/27448/

• Post #4280177

  GitHub launched stacked PRs today! https://github.blog/changelog/2026-07-30-stacked-pull-requests-are-now-in-public-preview/ Who's started using them already? Link me to your stack. 🥞

• Post #3967002

  Next week, my Microsoft colleague and I are running a livestream series showing how to ground AI apps in real data: "Microsoft IQ Deep Dive with Python" We'll cover: * Foundry IQ (Azure AI Search): create multi-source knowledge bases with hybrid search and re-ranking * Fabric IQ: query ontologies from MS Fabric * Work IQ: query M365 data like teams, chats, calendar events, etc 🗓️ July 28–30, 10 AM PT 👉🏼 Register: https://aka.ms/IQDeepDivePython/series

• Post #3341075

  You can now set a limit for the number of allowed open PRs for users without write access on GitHub repos: https://github.blog/changelog/2026-06-17-limit-open-pull-requests-for-users-without-write-access/

• Post #1883158

  We got a Copilot stuffie, and now my kid is walking "little baby Copilot" around in a stroller, saying "shh Copilot is crying, pobrecito Copilot." 😆

• Post #1256934

  Got a bunch of spam comments on a commit this morning. Time to shut that down! (GitHub > Settings > General > Commits)

• Post #1096891

  “the goal today is for you to understand data classes, because one day you’ll outgrow them. “ - Brett Slatkin #PyBay2025

• Post #903568

  Going to be on the BART for 4 hours today! First, it's down to San Jose for NVIDIA GTC: https://www.nvidia.com/gtc/ Then it's back up to SF for the DSpy meetup: https://luma.com/je6ewmkx

• Post #903567

  "OSS in the Age of AI: PyAI Panel discussion" https://pydantic.dev/articles/pyai-oss-panel

• Post #903566

  people are seriously sleeping on the cuteness of skipper butterflies! https://www.inaturalist.org/taxa/1081332-Lon-melane attached for evidence: a skipper in my garden today

• Post #903565

  Which Ollama SLMs are people using *successfully* with agent frameworks like Langchain/Pydantic-AI/Agent-framework/etc? I am trying qwen3.5:9b since that's best I can run on my 16GB RAM Mac, but it often outputs in chinese and forgets its original task.

• Post #903563

  You can now watch my talk from PyAI about improving MCP server tool schemas to increase agent reliability: https://www.youtube.com/watch?v=l0McGgmngeU Also check out the: * Slides: https://pamelafox.github.io/py-ai-mcp-tool-schemas/ * Repo: https://github.com/pamelafox/py-ai-mcp-tool-schemas * Write-up: https://blog.pamelafox.org/2026/03/do-stricter-mcp-tool-schemas-increase.html

• Post #903562

  "Mirage: The Illusion of Visual Understanding" https://arxiv.org/abs/2603.21687 New research shows that multimodal LLMs may base their answer/reasoning on a non-existent image. Includes a new eval framework B-Clean for filtering compromised questions in benchmarks.

• Post #903561

  uvicorn, why can't i quit you?!?!

• Post #903559

  FastMCP v3 has 5 ways of building MCP apps: 1. Prefab apps (common UI components) 2. FastMCPApp (UI composability) 3. Generative UI (LLM-generated) 4. Custom HTML (DIY HTML/CSS/JS) 5. App providers https://gofastmcp.com/apps/overview

• Post #903558

  mcp-server-diff: A GitHub action for detecting differences in MCP servers between versions. https://github.com/SamMorrowDrums/mcp-server-diff Good way to ensure all agent-facing MCP server changes are intentional!

• Post #903557

  I am so done with LLMs adding inline imports to my Python code. Who does that?? Adding this prompt to my global instructions to hopefully dissuade inline imports forever.

• Post #903556

  Who's coming to PyCon sprint days this year? Debating whether to attend - it makes a long trip away from home even longer, but I also hear it's really a great experience.