Staff Research Scientist - Post-Training for Agents
Core
Drive research in post-training and autonomous agents, advancing reinforcement learning, simulated environments, and synthetic data generation for complex real-world systems.
Role type
Staff Research Scientist (hands-on IC)
Builds
Simulated environments, synthetic data pipelines, evaluation frameworks, and scalable agent training capabilities for Datadog's observability and security platform.
Domain
AI Research / Reinforcement Learning / Autonomous Agents
Deliverable
production ML models
Required skills
Reinforcement learning, AI agents, post-training methods, generative modeling, simulated environments, synthetic data generation, evaluation frameworks, research program leadership, technical mentorship
Preferred skills
Publications at top-tier conferences (NeurIPS, ICLR, ICML), open-source contributions, cross-functional collaboration
Technologies
Reinforcement learning frameworks, simulation environments, synthetic data tools
Responsibilities
Set technical direction for research programs, own end-to-end research problems from framing to evaluation, design and advance RL approaches and training loops, build scalable training and evaluation infrastructure, mentor researchers and engineers, translate research into product capabilities
Seniority
Staff, hands-on IC with strategic influence
