Today
- October 9, 2025 Generally, I appreciate this call for precision from Jeremy Keith: When I talk about large language models, I make sure to call them large language models, not “AI”. I know it’s a lost battle, but the terminology matters to me. However, I want to offer a...
Yesterday
- This isn’t necessarily surprising, but it’s worth noting anyway. Claude Sonnet 4.5 is capable of building a full Datasette plugin now. I’ve seen models complete aspects of this in the …
This week
- OpenAI's dev day was today. While I wrote up a short summary of what was announced on bluesky, one of the major announcements was the AppSDK for ChatGPT. It looks like OpenAI plans to position ChatGPT as a platform for the future not unlike the Google Play and...
- Kicking the tires on an initial, naive agentic search with some thoughts on how it could be improved further...
- A fantastic deep dive into the seahorse emoji phenomena[1] was recently published by Theia[2]. It's engaging, well presented and worth reading. The post presents a case using meta-llama/Llama-3.3-70B-Instruct. However I wanted to verify this behavior with smaller models which...
Last week
- What’s the best way to get started building AI agent systems? There are countless frameworks for building AI agents available, such as CrewAI, LangGraph, and the OpenAI Agents SDK, and it can be overwhelming to choose one. On the other hand, Anthropic recommended starting with...
- AI code-writing tools make it possible for more people to write code than ever before. On my current team, I’ve seen PRs opened by engineers, designers, marketers, managers, and product people. There’s something special about empowering more people to build and fix things. That...
- In this post I explain why today’s humanoid robots will not learn how to be dexterous despite the hundreds of millions, or perhaps many billions of dollars, being donated by VCs and major tech companies to pay for their training. At the end of the post, after I have completed my...
- One frustrating pattern I’ve noticed in AI agents - at least in Claude Code, Codex and Copilot - is building automatic fallbacks. Suppose you ask Codex to build a system to automatically group pages in a wiki by topic. (This isn’t hypothetical, I just did this for EndlessWiki)....
- You are a software engineer. Don't become a prompt refiner....
Two weeks ago
- Service templates are a typical building block in the “golden paths” organisations build for their engineering teams, to make it easy to do the right thing. The templates are supposed to be the role models for all the services in the organisation, always representing the most up...
- edit: I temporarily disabled new page generation because of automated traffic, but I’m re-enabling it with a rate limit (and openai/gpt-oss-120b instead of Kimi-K2). I built an infinite, AI-generated wiki. You can try it out at endlesswiki.com! Why build an AI-generated wiki?...
- I gave a short address for the first AIE Paris conference organized by Koyeb!...
- Birgitta Böckeler examines the risk assessment around when to use vibe coding, using three dimensions of risk: Probability, Impact, and Detectability more…...
- I spent the last couple of weeks building an AI-driven spaced repetition app. You can try it out here. Spaced repetition Like many software engineering types who were teenagers in the early 2000s1, I’ve been interested in this for a long time. The main reason is that, unlike...
- Agents need tools they understand, like simple keyword search. They can reason about these tools, evaluate the results, refine, and iterate to deliver rather interesting results. But maybe with some caveats....
- TLDR: I built a simple & fun game using React Native, try it out on iPhone and iPad! In the Beginning…. My kids are 7 & 9 as I write this in 2025, but 6 years ago in 2019 they loved a simple little bubble popping app on my phone, and I thought it’d be … Continue reading Super...
Rows per page
