Overview # After coming across Ha & Schmidhuber’s 2018 paper World Models earlier this year, I have been working on building a “tiny world model” on a clone of Chrome’s Dino game. The final model is small: a variational autoencoder (VAE), a recurrent world mod…
NO THANK YOU.
A terrible email to receive, and likely the first of many.
A memory rule wasn't enough to stop Claude Opus 5 from turning quietly agreeable in deep sessions. Notes on the four-leg doctrine I put in place instead, and a July paper that reframes what I built.
From a "weird bug" to massive speedups
A rare opportunity to make a real impact in a fast-paced, high-visibility role.
LSPs for LLMs
Back in the dark ages of typing code into editors we were aided by squigglies under broken code, click-to-definitions links, and so on. That was powered by language servers and type checkers. Sever…
Building Autonomous Goal Loops That Deliver - JXØ
How to give a coding agent a real world, real demand, a protected measure, and enough memory to improve a complex feature.
Primer: Worktrees + Coding Agents
Coding agents and (Git) worktrees seem like a match made in heaven. However, there's a lot more to it than meets the eye. Over the last 12 months, I have learnt way too much about Git worktrees, and I think now is a good time to lay it all
I do not like to engage any form of gossips. I find it unproductive use of my time. If I have the chance, I rather use the time to read something or built so...
The Ultimate Dream
Coding is not solved. What problem solving actually is, why participation beats winning, and why writing specs is the ultimate accountability.
Shortlinks — Cailin Pitt
I built a personal shortlink service – cailin.link
Using AI to build is cool until you shut your brain off and let it decide what the product should be.
I had a random discussion in my Telegram group about AI. I noticed that some of them had really high token usage and very long sessions. After some more discussion, I realised that they had fairly complex setups, with lots of coordination between their agents.…
LLMs are making me lose my savviness
Things are getting boring - using LLMs is making software development not fun anymore
I never questioned why code should be highlighted until I tried to make my own theme.
Personal blog of Stephen Diehl - Software engineer writing about technology, programming, and the future
Just Ship | 0sql
Try to ship as often as possible.
Debian votes to allow 'responsible use of generative AI'
The results of the Debian general-resolution vote on the use of large language models have been [...]
GitHub - Lakr233/vphone-cli
Contribute to Lakr233/vphone-cli development by creating an account on GitHub.
Our decision to wind down our contract providing OpenAI models to Cursor following its acquisition by SpaceX.
Just a rumour of a bug is enough to find a security exploit these days
Thinking through how the conventional OSS security embargoes no longer buy us time, and what open source maintainers might do instead to respond