LLM Reinforcement Learning Framework Intern - 2027
Core
Intern working on NVIDIA Nemo RL to optimize large-scale post-training workloads and adopt SOTA RL algorithms.
Role type
Intern, IC machine-learning engineer (LLM RL)
Builds
NVIDIA Nemo RL framework
Domain
AI / Large Language Models / Reinforcement Learning
Deliverable
production ML models
Required skills
Python, C++, CUDA, PyTorch, RLHF, LLM training frameworks (FSDP/Megatron)
Preferred skills
VeRL, Miles, Slime, Nemo RL, agentic RL, inference engine development
Responsibilities
Optimize NVIDIA Nemo RL for large-scale post-training workloads, Adopt SOTA RL algorithms to Nemo RL, Improve RL rollout performance
