Prime Intellect
Applied Researcher - Agentic RL
$300K
San Francisco +3 · Full-time · Posted Oct 19, 2026
- Location
- San Francisco
New York
Remote
London - Visa sponsorship
- Visa Sponsorship
About the job
Conduct applied research on reinforcement learning and ML systems for long-horizon, tool-using and multi-agent systems, developing methods and infrastructure that are tested across real application domains.
What you'll do
- Develop RL and post-training methods, reusable environments, harnesses, rewards, and evaluations for agent systems.
- Run and analyze experiments, including identifying reward hacking, data leakage, harness behavior, and infrastructure artifacts.
- Build open-source systems and papers, and work with application teams, customers, users, and research partners.
What they ask for
- Hands-on experience in RL training, scalable ML systems, agent environments, evaluation, training or inference infrastructure, or research spanning methods and systems.
- Experience building and debugging model-training, post-training, agent, evaluation, or closely adjacent experimental loops.
- Strong experimental judgment and ability to attribute failures and revise approaches based on evidence.
- Strong Python skills and ability to build artifacts others can run, inspect, and extend.
Pay
$150-300k (base salary)
Tools
Python
AI focus
Reinforcement learning and post-training for long-horizon, tool-using, multi-agent, and memory-augmented AI systems; agentic evaluation, reward and credit assignment, and scalable training.
Model-written summary · the original posting is authoritative