Description
As a key player in the AI revolution, NVIDIA is pushing the boundaries of what's possible in accelerated computing. We are seeking an exceptionally dedicated LLM Reinforcement Learning Framework Engineer to join our ambitious team. This role is vital in advancing our large language model (LLM) capabilities, particularly in improving reasoning abilities for math, coding, and agentic AI.
The successful candidate will develop and deploy reinforcement learning algorithms for LLM post-training to improve reasoning and alignment. They will integrate RL components into NVIDIA's LLM training and serving stack with a cross-functional team of engineers and researchers. Additionally, they will craft and run experiments, evaluations, and debugging workflows to ensure robustness, scalability, and reliability in production.
To be successful in this role, you will have strong Python programming skills with production-quality PyTorch experience in multi-GPU and distributed training environments. You will also have hands-on experience with modern LLM frameworks such as NeMo RL, Megatron-LM, DeepSpeed, vLLM, TensorRT-LLM, or similar. Practical experience with reinforcement learning applied to LLMs or large-scale sequence models is a plus.
You will be familiar with async and distributed orchestration (e.g., asyncio, torch.distributed, Ray, or equivalent). You will have a solid foundation in probability, optimization, statistics, and deep learning. Understanding of GPU architecture and performance optimization is a strong plus.
Join us at NVIDIA and help build the next generation of reasoning-capable LLMs that make a lasting impact on the world!