Applied Research - Agentic RL

Role overview Research and build reinforcement learning methods and systems for agents that use tools, work across long tasks, coordinate with other agents, and learn from their experience. The work spans algorithms, environments, reward and credit assignment, evaluation, and scalable experiments, with an emphasis on methods that transfer across application areas and can be tested with external users and partners.

Responsibilities - Develop RL and post-training approaches for tool-using, multi-agent, memory-enabled, and long-horizon systems. - Build reusable environments, agent harnesses, reward signals, and evaluation methods. - Investigate how algorithm choices affect training and inference behavior through controlled experiments. - Diagnose apparent gains and failures by inspecting traces and checking for reward hacking, data leakage, or harness and infrastructure effects. - Create research artifacts and open-source systems that other teams can run, inspect, and extend. - Work with application teams, partners, and users to test methods against practical tasks while maintaining a research agenda.

Requirements - Hands-on experience in reinforcement learning, 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 similar experimental workflows. - Strong experimental judgment, including useful baselines, trace inspection, and clear attribution of failures. - Strong Python skills and ability to produce reusable, inspectable research software. - Ability to take ownership of ambiguous work and keep collaborators informed as it progresses. - Evidence of substantial work, such as publications, open-source systems, benchmarks, or shipped products; be ready to explain a difficult project and how evidence shaped your approach.

Benefits and work setup - Full-time, with work locations listed in San Francisco, New York, London, or remotely. - Listed cash compensation is $150,000–$300,000, plus equity incentives. - Visa sponsorship and relocation support, a professional development budget, and team off-sites and conference attendance are offered.