NVIDIA Shanghai LLM Reinforcement Learning Algorithm Engineer

[zoe1016aaa]

NVIDIA Shanghai is hiring for the position of LLM Reinforcement Learning Algorithm Engineer. As a leader in the global AI revolution, NVIDIA is committed to advancing accelerated computing. This role focuses on enhancing large language models in reasoning, math, coding, and agentic AI.

Location: Shanghai
Application email: [email protected] (Please use the subject line “Application – LLM RL Algorithm Engineer – Your Name”)

Key Responsibilities:

• Design and develop state-of-the-art reinforcement learning algorithms tailored for large language models (e.g., RLHF, RLAIF, curriculum RL).
• Collaborate with a world-class team of engineers and researchers to integrate RL algorithms into real-world applications and products.
• Utilize a strong background in mathematics and AI algorithms to improve LLM reasoning in complex problem-solving, math, coding, and agentic behavior.
• Manage the full cycle from idea generation, experimentation, evaluation to continuous refinement, ensuring stable and reliable model behavior.
• Contribute to NVIDIA’s mission of delivering industry-leading AI solutions and maintaining leadership in LLMs and accelerated computing.

Qualifications:

• Strong programming skills in C++ and Python.
• 3+ years of relevant industry experience.
• BS or MS (or equivalent experience) in Computer Science, Computer Engineering, Electrical Engineering, Mathematics, or related fields.
• Proven experience in reinforcement learning, ideally applied to large language models or sequential decision-making problems.
• Solid foundation in mathematics and AI algorithms, focusing on RL methods.
• Demonstrated history of applying RL algorithms in practical or production scenarios.
• Understanding of GPU architecture and CUDA is a strong plus.
• Excellent problem-solving skills and the ability to collaborate in a dynamic team.
• Genuine passion for innovation and pushing the frontier of AI.

If you fit this description, please send your CV along with representative projects, GitHub, or publications to [email protected]. Feel free to share this post with anyone who might be a great fit. Join NVIDIA and help