Nora Petrova is a Staff AI Researcher in London with a decade of engineering experience bridging full-stack development and machine learning. She has moved from hands-on software and DevOps roles into senior ML and safety research positions, currently working on activation steering and latent adversarial training to improve refusal behaviours in LLMs. Her recent work spans applied safety research at Prolific, LASR Labs and Apart Research, investigating which activation directions most influence model outputs using KL-divergence–based analysis. Nora combines product-focused engineering leadership (Signal AI, Primer.ai) with deep technical training in reinforcement learning and ML engineering from Udacity, and an academic grounding in physics and IT. Colleagues appreciate her rare mix of pragmatic production experience and rigorous model interpretability research.
10 years of coding experience
9 years of employment as a software developer
Bachelor of Science (BSc) Physics, Bachelor of Science (BSc) Physics at The Open University
Deep Reinforcement Learning Artificial Intelligence, Deep Reinforcement Learning Artificial Intelligence at Udacity
BA Information Technology, BA Information Technology at Molde University College
Contributions:12 commits, 3 PRs, 17 pushes in 13 days
couchdbpermissionsdatabasestorage
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.
Request Free Trial
Nora Petrova - Staff AI Researcher at Apart Research