Build security environments, train adversarial agents, and evaluate frontier AI systems. Location: Tübingen / Hybrid Employment: Full-time The role You will work across AI security, machine learning, and research engineering. You will design reinforcement-learning environments, train adversarial agents, and build evaluations for capable AI systems. What you’ll work on • Design security environments for training and evaluating AI agents • Post-train agents using reinforcement learning and develop adaptive red-teaming methods • Evaluate frontier models and agentic systems • Turn research results into benchmarks, training methods, and tools What we’re looking for • A strong track record in machine learning, security, or systems • Strong software and machine-learning engineering fundamentals • Experience with LLMs, agentic systems, or reinforcement learning • Ability to run careful experiments and communicate results clearly Useful experience • AI security, red teaming, or adversarial machine learning • Reinforcement learning or post-training for language models • Research engineering or production ML systems • Published research or substantial open technical work What we offer • Competitive compensation with equity • Flexible hours and a remote-friendly setup • Vacation, learning, and conference budget • Research freedom in a high-trust environment Work with us Submit your CV, a short note, and links to work that shows how you think and build using the application form.