Yesterday
- I mostly link to written material here, but I’ve recently listened to two excellent podcasts that I can recommend. Anyone who regularly reads these fragments knows that I’m a big fan of Simon Willison, his (also very fragmentary) posts have earned a regular spot in my RSS...
- Meta announced Muse Spark today, their first model release since Llama 4 almost exactly a year ago. It’s hosted, not open weights, and the API is currently “a private API …
This week
- Rahul Garg finishes his series on reducing the friction in AI-Assisted Development. He proposes a structured feedback practice that harvests learnings from AI sessions and feeds them back into the team's shared artifacts, turning individual experience into collective...
- Running multiple agents in parallel is not just a question of throughput. It is a new kind of cognitive labor that requires managing multiple mental models, continuous judgment calls, and an ambient anxiety tax...
- Anthropic didn’t release their latest model, Claude Mythos (system card PDF), today. They have instead made it available to a very restricted set of preview partners under their newly announced …
- April 7, 2026 It’s a small thing, yet it says a lot, that OpenAI’s Industrial Policy for the Intelligence Age is presented only as a PDF that looks terrible, with amateurish justification and a footer image that’s too lo-res and blurry for clear printing. I’ll note also...
- Google's Gemma 4 family of open models is now available through the Gemini API and Google AI Studio. Built from the same research behind Gemini 3, these models bring advanced reasoning, native function calling, multimodal understanding, and 256K context windows to an open,...
- April 6, 2026 A new edition of my pop-up AI newsletter just landed: where is it like to be a language model? The discussion here is bolstered by an actual experiment, a programmatic probe of many language models. It was my first time doing something like that — fun! The...
- Did you know that Jesus gave advice about prototyping with an LLM? Here’s Luke 14:28-30: Suppose one of you wants to build a tower. Won’t you first sit down and estimate the cost to see if you have enough money to complete it? For if you lay the foundation and are not able to...
- With enough work you can build a search system with just grep. But it’s not for the faint of heart....
- I like reading system prompts, either when they’re published as part of open-source software, exfiltrated via crafty prompting, explicitly shared, or (in the case of last week) accidentally leaked. They’re often the best manual for how an app is intended to work. We’ve touched...
- AI Assisted/Agentic programming are pretty common place at this point. The growing sentiment seems to be that if you can't find some sort of benefit in your workflow, it's more of a skill issue than a problem with the tools. Whether you believe this to...
- You may be seeing posts claiming METR’s widely-cited 2025 study has been followed up with new research showing an 18% productivity boost. That’s not what the article says. METR: We are Changing our Developer Productivity Experiment Design In 2025, METR found experienced...
- Back in 1985, computer scientist Peter Naur wrote “Programming as Theory Building”. According to Naur - and I agree with him - the core output of software engineers is not the program itself, but the theory of how the program works. In other words, the knowledge inside the...
- I have no connection to the authors of the Superpowers plugin for Claude Code, but I have been raving about it to everyone I talk to. Using Claude Code with Superpowers is so much more productive and the features it builds are so much more correct than with stock Claude Code. I...
- I was a guest on Lenny Rachitsky’s podcast, in a new episode titled An AI state of the union: We’ve passed the inflection point, dark factories are coming, and automation …
Last week
- Yesterday we talked about how cheap code is fueling an era of idiosyncratic tooling, and previously we’ve talked about the rise of spec driven development. In that second piece, we ran through some of the initial examples of spec driven development with agents: By far, the...
- As we see LLMs churn out scads of code, folks have increasingly turned to Cognitive Debt as a metaphor for capturing how a team can lose understanding of what a system does. Margaret-Anne Storey thinks a good way of thinking about these problems is to consider three layers of...
- Last month Birgitta Böckeler wrote some initial thoughts about the recently developed notion of Harness Engineering. She's been researching and thinking more about this in the weeks since and has now written a thoughtful mental model for understanding harness engineering that we...
- CircleCI’s 2026 State of Software Delivery report has two findings that are already travelling: AI is meaningfully boosting software delivery, but only 1 in 20 teams are capturing that benefit. Both claims are more uncertain than the report suggests, for different reasons. What...
- AI coding assistants respond to whoever is prompting, and the quality of what they produce depends on how well the prompter articulates team standards. Rahul Garg proposes treating the instructions that govern AI interactions (generation, refactoring, security, review) as...
- One of the folks who joined my presenter workshop last week (which was awesome, by the way!) emailed me this morning with a follow-up question: I first saw you speak at DDD South-West in Bristol (it was the “There’s No Such Thing as Plain Text” talk), and what stuck with me was...
- I used to post a lot of stuff on Twitter when it was good. It’s all still on there somewhere, and I have archives of it all, but I’m reposting this thread, which I originally wrote back in 2019. Partly because I dug it out of some archives tonight and thought “hey, that’s not...
- Trip Venturella released Mr. Chatterbox, a language model trained entirely on out-of-copyright text from the British Library. Here’s how he describes it: Mr. Chatterbox is a language model trained entirely …
Rows per page
