Artem Karpov is a senior ML and AI safety research engineer with 13 years of software engineering experience and a recent focus on monitoring and controlling language model agents. He investigates subtle failure modes like steganographic collusion and hidden chain-of-thought encodings that can undermine safe LLM deployments, with papers at AAAI, ICLR, and NeurIPS and funding from major nonprofits. Since pivoting to AI safety in 2022 he has combined deep ML programs (MATS, ARENA, MLSS, Apart Fellowship) with hands-on engineering—building LM agent evaluations for the UK AI Security Institute and participating in METR baselines. Earlier in his career he designed emergency-response systems, contributed to .NET Core ecosystems, and delivered high-throughput backend services used by millions, giving him a rare blend of production-grade engineering and research rigor. Based in Istanbul, he brings applied-math training to practical ML safety problems and a demonstrated knack for turning research into reproducible evaluations and deployed experiments.
13 years of coding experience
15 years of employment as a software developer
ML Engineering for AI Safety research, ML Engineering for AI Safety research at ARENA (Alignment Research Engineer Accelerator)
Bachelor's degree, Applied Mathematics, Bachelor's degree, Applied Mathematics at Far Eastern State Transport University
ML Engineering in AI Safety, ML Engineering in AI Safety at Machine Learning Safety Scholars
This repo contains the .NET Core foundational libraries, called CoreFX. It includes classes for collections, file systems, console, XML, async and many others. We welcome contributions.
Contributions:77 pushes, 11 branches in 2 years 3 months
Space to practice algorithms and other code problems
Contributions:90 pushes, 1 branch in 6 years
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.