HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Thu, 06 Nov 2025 18:30:42 GMT access-control-allow-origin: * etag: W/"690ce952-102a" expires: Tue, 30 Dec 2025 08:38:12 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: 9C96:234FE9:9E7EA2:B1DDA1:69538D1B accept-ranges: bytes age: 0 date: Tue, 30 Dec 2025 08:28:12 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210024-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767083292.485096,VS0,VE213 vary: Accept-Encoding x-fastly-request-id: 844eef039973c3444868b91fbb60b3236c35709e content-length: 1631

I am a senior research scientist at NVIDIA's Applied Deep Learning Research team working on reinforcement learning for LLMs. Before, I was a research scientist at Isomorphic Labs working on AI-first drug discovery. Prior to that, I was a PhD student at the University of Oxford working on multitask reinforcement learning with graph-based state-action representation. My scientific advisor was Shimon Whiteson. I'm also known as yobibyte. Ping me if you are interested in my work. The head is clickable, btw.

links

• RSS feed for this website
• github
• twitter
• google scholar
• CV

machine learning

• ICML 2024 compressed
• ICLR 2024 compressed
• Compressor: my LLM-based scientific papers summarisation project.
• arxiv compressed
• Learning on Graphs 2023 compressed
• Neurips 2023 compressed
• My machine learning papers summaries
• an overview of the evaluation procedures for the Atari 2600 domain
• all you need is a good init
• ongoing survey on starcraft research
• reinforcement learning summer school 2017

programming

• self-reliant programmer
• reading safetensors in zig
• thoughts on vibecoding

setup

• kitty productivity setup
• why I got rid of all my neovim plugins
• I got rid of all neovim plugins
• my 90% terminal and mouseless setup (i3, w3m, neovim, tmux)
• framework, arch and kernel update
• my neovim setup
• notebooks are McDonalds of code
• Newsboat hack

grad school

• DPhil Grind book
• Productive grad school
• NVIDIA 2019 Internship Post Mortem
• Microsoft Research Cambridge Visit
• art of juggling and reinforcement learning
• gonzo workstation setup

misc

• mathclub
• advent of terminal
• hard tech is the way to go
• John Blow on how to Deal with Mental Health Issues
• how I taught my son play chess
• Shit Non-Proliferation Manifesto
• Cholesky Decomposition

---