Lukasz Olejnik
I'm open to consulting, full-time or part-time work across research, engineering, security, AI, regulation and policy. AI-first work very much in scope. This can include commissioned research, adversarial security/privacy review, AI and agent-system, technical analysis, and policy/regulatory work. me@lukaszolejnik.com
Lukasz Olejnik
I'm proud of my recent work. 1) AI-powered autonomous influence operations, 2) AI-assisted cryptanalysis, and 3) a concrete break of a proposed FHE construction. - IO Factory - an agentic environment for simulating AI-enabled influence campaigns at scale, including runs with up to 100,000 agents. I learned a lot here about building AI-powered systems, orchestration, simulation, and using AI coding and testing assistants while keeping experiments traceable and reproducible. URL: https://arxiv.org/abs/2608.10920 - AI Grinding for Fun and Cryptanalysis - a framework for automated AI-assisted cryptanalysis. Agents generate, test, and refine attack hypotheses at scale. The workflow uncovered multiple security failures in published cryptographic constructions. https://arxiv.org/abs/2608.21986 - Key Recovery from Residue-Confined Errors in the CRT-RLWE Construction - focused cryptanalysis of a proposed FHE construction. https://eprint.iacr.org/2026/1734