my new favorite OpenAI Codex use-case - see prod issue starting on a specific date. - ask Codex: “which code change could’ve triggered this?” - walk away. - come back to a list of suspects and evidence peak debugging ROI

wtf anthropic dumped 17 new videos (8 hours worth of content) and some of them are pretty good: > building ai agents with claude > claude plays pokemon > claude code best practices > prompting for agents > how students build with claude > vibe coding in prod

worst graph crime ive ever seen, what the fuck is this

Google just released LangExtract Python library. It can extract structured data from unstructured docs with precise sources in just a few lines of code. 100% Opensource.

Cognition invited SF’s most ambitious builders to push GPT-5, Claude Opus, and Gemini Pro to their limits And they invited Andrej Karpathy to pick the very best. Here are the top demos from the Cognition Applied AI Hackathon (🧵):

Continuing the journey of optimal LLM-assisted coding experience. In particular, I find that instead of narrowing in on a perfect one thing my usage is increasingly diversifying across a few workflows that I "stitch up" the pros/cons of: Personally the bread & butter (~75%?) of

А литкодные задачки инженер это родивший все прорешал

this is who rejected your PR

OpenAI just released the largest-ever report on how consumers are using ChatGPT. It looks at millions of (anonymized) messages over the last 3 years to analyze usage evolution over time. Key takeaways 👇

You're in a ML Engineer interview at Perplexity, and the interviewer asks: "Your RAG system is hallucinating in production. How do you diagnose what's broken - the retriever or the generator?" Here's how you can answer:

Today we are launching my favorite feature of ChatGPT so far, called Pulse. It is initially available to Pro subscribers. Pulse works for you overnight, and keeps thinking about your interests, your connected data, your recent chats, and more. Every morning, you get a

very important work on a new eval

Simon Willison Have you looked at Chat Debug View for GitHub Copilot? It shows every request that goes to the model - I like seeing the concatenation of VS Code system prompt + my custom instructions + tool definitions.

Cursor CPO Sualeh Asif Turbopuffer CEO Simon Eskildsen Notion AI Engineer Abhishek Modi Braintrust CEO Ankur Goyal Great discussion on all things Semantic Search and honestly the hard parts of AI Engineering some spicy things said - “in my entire career nobody uses knowledge graphs” -

It only took 13 years, but dark mode is finally here sebastianraschka.com/blog/2021/dl-c…

Crying in RAG

The new Tesla AI5 chip In-house design 40× faster 8× compute 9× memory 5× bandwidth Code paths shrunk to ~5 10× cheaper per inference than Nvidia 3× more efficient per watt Built by Samsung with lithography by TSMC all performed in Texas and Arizona. Production 2026. It is a

I took delivery of a beautiful new shiny HW4 Tesla Model X today, so I immediately took it out for an FSD test drive, a bit like I used to do almost daily for 5 years. Basically... I'm amazed - it drives really, really well, smooth, confident, noticeably better than what I'm used

I’m starting to get into a habit of reading everything (blogs, articles, book chapters,…) with LLMs. Usually pass 1 is manual, then pass 2 “explain/summarize”, pass 3 Q&A. I usually end up with a better/deeper understanding than if I moved on. Growing to among top use cases. On

🚀 Launching DeepSeek-V3.2 & DeepSeek-V3.2-Speciale — Reasoning-first models built for agents! 🔹 DeepSeek-V3.2: Official successor to V3.2-Exp. Now live on App, Web & API. 🔹 DeepSeek-V3.2-Speciale: Pushing the boundaries of reasoning capabilities. API-only for now. 📄 Tech