Researcher - Reinforcement Learning
Core
Enabling Large Language Models (LLMs) to learn from experience via agentic reinforcement learning, focusing on post-training paradigms (RLHF, GRPO) and tool-using agents.
Role type
Researcher - Reinforcement Learning (LLM post-training)
Builds
Training and evaluation pipelines for reasoning-enhanced LLMs and tool-using agents
Domain
Artificial Intelligence / Large Language Models / Reinforcement Learning
Deliverable
production ML models
Required skills
Deep learning (Transformers), Reinforcement Learning, Python, PyTorch, Distributed training frameworks (DeepSpeed, Megatron), LLM post-training (RLHF, GRPO, SFT, LoRA, MoE)
Preferred skills
Multi-agent RL, Tool-use / browser/coding agents, Self-supervised learning
Technologies
PyTorch, DeepSpeed, Megatron
Responsibilities
Implementing and evaluating new training and evaluation pipelines for reasoning-enhanced LLMs, Scaling experiments on large GPU clusters, Contributing to scientific insights and publications
Seniority
Senior, Researcher (PhD or Master's with comparable experience)