Frontier LLMs continue to get more and more powerful, exactly as expected . Yet despite their increased usefulness, a camp of AI skeptics and never-LLMers is emerging - Zig goes as far as a Strict No LLM / No AI Policy ! There was some merit to this, early LLMs genuinely needed close supervision and technical expertise to produce high-quality code, and overly enthusiastic early adopters produced a…
I'm very excited to finally be able to share more details about how AlphaProof works! AlphaProof is the system that we used to discover the Lean proofs for the International Mathematical Olympiad 2024 , reaching silver medal performance. Our full paper Olympiad-Level Formal Mathematical Reasoning with Reinforcement Learning has now been published in Nature. : The work described was done from the…
This week I had the pleasure of chatting with Matt Turck from the MAD Podcast , and it was a fun one! We talked trends in AI, RL and why it unlocks Agents, scaling and so much more. You can find the episode on your podcast app of choice, or on YouTube: For the curious, some links to what we talked about, and further reading: my recent blog post, Failing to Understand the Exponential Move 37, as…
Python 3.14 is released and I’m extremely exited that free-threading is now officially supported! Finally no more GIL, finally able to use all the hundreds of cores on modern servers! Or at least in theory, we'll see how it holds up in practice. I'm also quite excited for template string literals (t-strings). Better HTML templates? Direct SQL interpolation? I'm looking forward to what the…
The current discourse around AI progress and a supposed “ bubble ” reminds me a lot of the early weeks of the Covid-19 pandemic. Long after the timing and scale of the coming global pandemic was obvious from extrapolating the exponential trends, politicians, journalists and most public commentators kept treating it as a remote possibility or a localized phenomenon. Something similarly bizarre is…
Good advice should rule out large parts of the possibility space - if you don't find yourself disagreeing with the advice or changing what you are doing, then the advice is just generic fluff. Advice should be based on experience. Not necessarily your own, observing others can also make good advice, and it scales a lot better. Good advice is contentious or not widely known. If everyone agrees with…
A timely topic, as I'm writing this on the plane back from my latest trip: When you go on holiday, don’t check your work email, disable all notifications, just disconnect completely. Not just for your own mental health - many people have covered that extensively - but also to test the resilience of your work projects. It’s easy to argue about bus factors , but nothing clarifies the situation…
After almost 18 years with my old domain ( furidamu.org , acquired August 2007!), I found a new domain that should be easier to remember and spell: julian.ac I've migrated my blog to the new domain, but I plan to keep the old domain and URLs active indefinitely, with redirects to the corresponding new address. Please let me know if you run into any issues! I was also tempted by a second domain,…
While migrating my blog to my new domain ( julian.ac !), I realized that the service I had been using for comments had not been maintained in a while, and did not support easy migration to a new domain. I've been very impressed with the simplicity and reliability of S3 in the >10 years I've been using it to host this blog, so why not use the same setup for comments? It's not like I need to support…
This morning I got nerd-sniped by Why your website should be under 14kB in size . I had already optimized my blog several times previously, most recently writing my own static site generator in Rust , but according to the Cloudflare URL Scanner my front page was still a shocking 242 kB ! How could this be? Turns out 124 kB were consumed just by highlight.js , and a further 75 kB by a single…
Estimates for the dynamic range (the maximum perceivable contrast between light and dark) of human eyes vary depending on the source, from 1:100 to 1:1000 or more for static contrast. Dynamic contrast, where the eye moves around the scene and the iris adjusts in size is many orders of magnitude larger - with sufficient adaption time we can switch between bright daylight ( 10 8 c d / m 2 ) and…
I mentioned in my previous post a lock contention issue when accessing the same list from different threads, leading to worse performance the more threads were used. Turns out this is fixed in Python 3.14.0a7 already! Unfortunately this version doesn't (yet) seem available from uv (which is why I hadn't originally tried it), but it can be downloaded and installed manually from the Python website…
As of version 3.13, Python is finally getting support for real multi-threading that is not hobbled by the GIL : experimental support for free threading ! This means multiple Python threads can now execute at the same time, without needing to compete for the global interpreter lock. In modern machines with hundreds of cores this can speed up CPU-bound programs by two orders of magnitude! Even in ML…
After 10 years of writing C++ while at Google, I’ve really enjoyed writing Rust over the last 6 months at Anthropic . It’s truly a breath of fresh air - modern APIs, confidence that if my code compiles it will work as expected, easy dependency management, a great linter. I say that as someone deeply familiar with modern high-performance C++, having touched around 1 million lines of C++ in various…
I'm excited to announce that I'm joining Anthropic , starting this week! I'm honoured to be joining such a capable and kind group of people, who I've long admired - Claude was the first LLM that I found myself consistently using. Recently I've been really blown away by Artifacts and Computer Use , and the consistent improvements in Claude's skills. I've also been incredibly lucky to be part of an…
Recently I decided to build a NAS to store my RAW photos as well as video content, and to use as a backup target for my MacBook. I wanted something that I could use directly with low latency for interactive editing over 10 GBit ethernet, so spinning disks were out. However, building an all SSD solution turned out to be surprisingly tricky - the old SATA standard is much too slow for modern SSDs,…
Recently I was asked which habits and practices were the secret to my productivity. I figured my answers might be of wider interest. Let me know what works for you! Why should you listen to my advice? Aside from my papers , in each year of the past decade I've authored ~1k pull requests with ~200k LoC of deltas, with a similar number reviewed: You can tell when I'm on holiday :) Learning My first…
I was asked some questions about the development of AlphaGo and figured my answers might be of wider interest, so here we go: How low-level does the code need to be in order to get the maximum performance both in training and, possibly, live game? To maximize playing strength it is necessary to balance maximum utilization of the accelerators (GPUs or TPUs) and focusing on exploring and evaluating…
Once again I've migrated this blog to a new backend - originally I had been using Octopress, then later I migrated to Pelican after I got fed up with the complicated setup and dependencies. Well, recently I had issues getting Pelican to work after some updates, so I decided to bite the bullet and write my own small static site generator: 開板 . Really it's just an excuse to play around with Rust and…
In the conventional Turing test (aka imitation game), an investigator tries to distinguish between a human and a computer solely by interacting with them. This is an interesting setup and has inspired much research, but it doesn't immediately translate into practical usefulness - a computer system may pass as human, but may still not be able to help me accomplish any task. Instead, I find I'm…