What Resources Are Available on Arthur Douillard's Website?

arthurdouillard.com is the personal research site of Arthur Douillard, a researcher at DeepMind in London. According to the site's own description, his work covers RL for scaling intelligence per FLOPs on Gemini, earlier work on AI for systems with distributed resilient pre-training, and a PhD on continual learning for computer vision. The site itself is organized around three resource areas: Research, Tools, and DeepCourse. If you are looking for his publications, software, or teaching material, those are the sections to check.

Research

The Research section is where his publications and projects are collected. Based on the site description, the topics you can expect to find there span:

• Reinforcement learning aimed at scaling intelligence per FLOPs, in the context of Gemini
• AI for systems, specifically distributed and resilient pre-training
• Continual learning for computer vision, the subject of his PhD

This is the section to use if you want to read his papers or follow a specific line of work rather than browse everything at once.

Tools

The Tools section lists software or utilities he has released. The site description does not enumerate individual tools, so treat this section as the place to look for code and practical artifacts tied to his research. If you arrived from a paper and want the accompanying implementation, check here before searching elsewhere.

DeepCourse

DeepCourse is an educational resource hosted on the site. It is listed alongside Research and Tools as one of the three main areas, which makes it the entry point for anyone who wants structured learning material rather than papers or code. The site description does not detail its curriculum, so open it directly to see the current scope and format.

How to Navigate the Site

The site is small and its structure maps directly onto the three areas above:

1. Open arthurdouillard.com.
2. Use the top-level navigation to pick Research, Tools, or DeepCourse depending on what you need.
3. Within Research, move to the specific topic or publication you are after; within Tools, open the individual tool; within DeepCourse, follow the course material in order.

Which Section Should You Use?

One practical note: because the site is a personal page rather than a large platform, the fastest route is usually to go straight to the section matching your intent instead of browsing the homepage in full. If a specific paper or tool is not listed under Research or Tools, it is likely not hosted here.