Datadog
Staff Research Scientist - Post-Training for Agents
$320K–$400K
New York, New York, USA +2 · Staff · Full-time · Posted Oct 7, 2026
- Location
- New York, New York, USA
Pittsburgh, Pennsylvania, USA
New York, NY, United States - Education
- PhD in Computer Science, Machine Learning, or a related field, or equivalent experience.
About the job
Leads hands-on research and development in post-training and reinforcement learning for autonomous agents, building training environments and evaluation methods and translating advances into Datadog products.
What you'll do
- Own research from problem framing through experimentation, model development, and evaluation; set technical direction for research programs.
- Develop reinforcement learning methods, agent training loops, simulated environments, synthetic data approaches, and evaluation frameworks.
- Mentor researchers and engineers, collaborate with Product and Engineering, and contribute to publications, conferences, and open-source artifacts.
What they ask for
- PhD in Computer Science, Machine Learning, or a related field, or equivalent experience.
- Deep expertise in reinforcement learning, AI agents, post-training, or generative modeling.
- Extensive hands-on experience building reinforcement learning systems, agent training methods, simulated environments, synthetic data pipelines, or evaluation frameworks.
- Track record of research impact through publications, model or system contributions, research artifacts, or equivalent technical achievements.
Pay
$320,000 — $400,000 USD (yearly salary)
Tools
reinforcement learning, simulated environments, synthetic data generation, evaluation frameworks, NeurIPS, ICLR, ICML
AI focus
Research and develop post-training methods, reinforcement learning, synthetic data, simulated environments, and evaluation for autonomous agents operating in complex systems.
Model-written summary · the original posting is authoritative