Robert Kirk is a research scientist at the UK AI Safety Institute focused on safeguards and reducing catastrophic AI risk, drawing on a PhD from UCL in foundational AI where he studied LLM safety, RLHF and generalisation in reinforcement learning. He combines deep research experience with two years of prior software engineering, shipping backend systems in Python/Postgres, front-end React apps, and cloud infrastructure on k8s and AWS. His work includes industry research at Meta on RL vs imitation fine-tuning for LLMs and hands-on development of reward models and distributed training pipelines. Comfortable moving between theory and production, he has facilitated AGI safety education and reformed operational processes in student governance, showing appetite for practical governance as well as technical rigor. Based in London, he brings ten years of technical experience and a rare blend of mechanistic interpretability, fine-tuning expertise, and production engineering chops. An understated strength is his track record of translating safety research into reproducible, deployable tooling and training workflows.
Contributions:12 PRs, 45 pushes, 3 branches in 4 years 8 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.