Senior AI Researcher

The Senior AI Researcher leads post-training evaluation, red-teaming, and reinforcement learning gym audits on open-weight models. The role involves establishing rigorous benchmarking methodologies, evaluating large language models against complex threats like Indirect Prompt Injections, and constructing post-training evaluation pipelines to measure frontier-level security capabilities. This position focuses on executing post-training runs, analyzing reward diagnostics, auditing tasks and environments, and integrating dockerized environments into internal training frameworks.