Research Scientist / Engineer – Reinforcement Learning Infrastructure

Highlights

• Build the systems that make reinforcement learning work at frontier scale.
• Turn model behavior into learning signal via reward and verification systems.
• Design agentic environments scalable to millions of episodes.

Ideal candidate

A systems-oriented engineer who has lived through large-scale RL post-training and thrives debugging complex, concurrent training loops to turn capable models into useful ones.

Requirements

• Hands-on experience post-training LLMs with RL (PPO/GRPO-family, RLHF, RLVR) at meaningful scale
• Extensive distributed PyTorch training and parallelism (FSDP, Tensor/Pipeline/Expert Parallel) for foundation models
• Experience building RL environments, reward functions, verifiers, or evaluation harnesses for LLM agents
• Deep familiarity with RL post-training frameworks (veRL, OpenRLHF, TRL, Ray) and rollout inference engines
• Strong understanding of GPU clusters, networking, and communication libraries under mixed workloads