Deep Learning Solution Architect - RL and Post-training
Core
Designing and optimizing Reinforcement Learning (RL) algorithms and infrastructure for Large Language Models (LLMs) and multimodal models, enabling open-source model support, and providing technical guidance to customers on integrating NVIDIA RL technologies.
Role type
Senior IC Deep Learning Solution Architect (RL and Post-training)
Builds
Production-ready RL frameworks, optimized training pipelines, and reusable toolchains for LLMs and multimodal models.
Domain
Artificial Intelligence, Reinforcement Learning, Large Language Models, High-Performance Computing (HPC)
Deliverable
production ML models
Required skills
Reinforcement Learning, Large Language Model training, multimodal learning, PyTorch, distributed training, task orchestration, evaluation pipelines
Preferred skills
RLHF, GRPO, DPO, alignment and post-training methods, scale-out HPC/cloud architectures, CUDA optimization, GPU performance tuning, agentic AI systems, code generation models, multimodal RL
Technologies
PyTorch, NVIDIA RL frameworks, CUDA
Responsibilities
Drive development and optimization of RL algorithms and infrastructure for LLMs and multimodal models; Develop and enable prompt open source models support in NVIDIA RL and post-training frameworks; Collaborate with internal research and engineering teams to adapt and validate state-of-the-art RL methods; Improve RL initiatives and engagements with customers; Develop and maintain reusable toolchains, experiment management workflows, and technical documentation.
Seniority
Senior, hands-on IC
