Latest / September 8, 2026
10 posts from engineering leaders published September 8, 2026, by Simon Willison, Jamie Tanna, Jeff Geerling and 5 more.
On the Navier–Stokes Millennium Prize Problem introduces an impressive result from OpenAI, who used an unreleased model to produce a resolution to the Navier–Stokes existence and smoothness problem , one of the seven Millennium Prize Problems that have been subject to a…
Introducing ChatGPT Images 2.5 OpenAI's image generation models are apparently used "more than 3 billion images across ChatGPT Images and the GPT‑Image models in the API". This latest release improves their instruction-following ability across multiple turns, responds faster,…
🤖 This post includes some LLM-derived content 🤖 In a lot of organisations, you'll find a fairly consistent structure in teams' (GitOps) repositories, where they set up a branch per deployable environment, such as: * dev * staging * prod Within this structure, the approach for…
UPSes are typically the most boring part of your homelab, but they're important. One of their main jobs is to give servers time to shut down safely if the power goes out. Another job is to condition your power so your servers can run smoothly. To do both of those things well ,…
Christian Catalini says we’re in a situation where we are vastly reducing the cost of generating things, but not the cost of verifying them: . This explains why the first major AI products appeared in chat, image generation, and code assistance. Not because these were the…
A GPU burns 700 watts and hundreds of thousands of tries to learn Pong. A human brain does it on 20 watts in a handful of attempts. So what can the oldest architecture in computing teach us about where agents and memory are headed?
Like me, Sumeet Gayathri Moghe is tired of poor presentations with bad slide decks. He's started to write a series of posts on how to avoid these calamities, beginning with a post that questions whether a presentation is needed at all. more…
I watched an IC have a lightbulb moment: they’re a manager now. Running agents in parallel, giving direction, deciding what’s worth doing. Being responsible for what ships, without doing it all themself. My job is giving them the pieces of management that are useful right now.…
Being technical enough has very little to do with how much code you write.
Fuck 1 it, we’re doing consumer hardware for ourselves now and getting fucked by economies of scale, I guess. This all started with me buying some Olimex Bluetooth proxies for Home Assistant. Then I started looking at the components and wondering if I could get better range. Now…