Taken from the location the company listed. Confirm in the posting before you apply.
What the job is
Designs and builds automated verification pipelines and RL environments that turn real software vulnerabilities into pass/fail reward signals for AI models.
What they ask for
Understanding of RL training workflows for LLMs, including execution-based feedback or RLVR
Proficiency in Python and low-level systems programming in C
Solid understanding of software vulnerabilities, binary exploitation, fuzzing, or program analysis
Experience with DevOps pipelines and reproducible builds (Docker, BuildKit, Nix) and Linux low-level debugging
Experience working with or building benchmark environments such as CTFs, SWE-bench, or execution sandboxes