Director of Post-Training Research

Job Summary

Leading the development of advanced AI systems in cybersecurity, the full-time remote Director of Post-Training Research will own the post-training pipeline for security-domain AI, manage a team of research scientists and engineers, and drive experimental work on complex problems while ensuring the integration of post-training and agentic research.

Key responsibilities

• Own and drive the full post-training pipeline for security-domain AI, setting research priorities and leading experimental work
• Build and maintain agent-RL training environments, contributing to environment design and reward shaping for realistic cyber workflows
• Develop evaluation methodologies for the agentic stack, ensuring reliable performance across security workflows and defining benchmarks for success

Required qualifications

• MS or PhD in computer science, machine learning, or a related quantitative discipline
• 8+ years of experience in ML research or engineering, with significant expertise in large language model post-training
• Hands-on experience with SFT data pipelines, RLHF/RLAIF, and reward model design
• Demonstrated ability to design and build agentic system harnesses for LLM-based agents
• Proven track record of leading high-velocity research programs and growing research teams while remaining an active technical contributor