Reinforcement Learning Engineer

Job Summary

Working remotely as a full-time Reinforcement Learning Engineer, the successful candidate will design and scale automated verification pipelines that transform real-world software vulnerabilities into deterministic reward functions, bridging low-level security analysis with modern LLM reasoning models.

Key responsibilities

• Design, build, and deploy high-throughput RLVR environments to evaluate LLM action sequences against deterministic outcomes
• Develop automated test harnesses and verification engines to convert vulnerability research into binary pass/fail reward signals
• Collaborate with researchers to define benchmark formats, observation spaces, and evaluation metrics for cybersecurity tasks

Required qualifications

• Understanding of RL training workflows, particularly Reinforcement Learning from Verifiable Rewards (RLVR)
• Proficiency in Python and low-level systems programming in C, with Rust experience being a plus
• Solid understanding of software vulnerabilities and fuzzing methodologies
• Experience with DevOps pipelines and reproducible builds using Docker or similar technologies
• Experience building benchmark environments for cybersecurity challenges