Senior AI Systems and Algorithms Engineer

NVIDIA•Santa Clara, CA

2d•Hybrid

About The Position

NVIDIA is seeking a Senior GenAI Algorithms Engineer to advance foundation model development, training, and deployment. The role spans large-scale data preparation, pre-training, post-training, inference optimization, and framework development, with work at the intersection of distributed training, reinforcement learning for LLMs and VLMs, model efficiency, multimodal AI, and open-source AI infrastructure. The engineer will collaborate with research, product, and infrastructure teams to design algorithms, optimize systems, and contribute to NVIDIA's open-source AI stack, including Megatron-LM, Megatron Bridge, and NeMo-RL. The role includes designing scalable systems for preparing high-quality multimodal datasets, improving the scalability, efficiency, and cost of large-scale training, advancing inference performance and deployment efficiency across cloud and edge platforms, and developing reusable infrastructure and model support for NVIDIA's open-source GenAI training platform.

Requirements

MS or Ph.D. in Computer Science, AI, Applied Mathematics, or a related field, or equivalent experience.
5+ years of relevant industry experience.
Strong foundation in machine learning, deep learning, and optimization.
Excellent software engineering skills, including Python and PyTorch.
Experience building high-performance software for large-scale AI systems.
Strong analytical, debugging, and performance optimization skills.
Excellent communication and collaboration skills.

Nice To Haves

Experience with distributed training at scale using frameworks such as Megatron-LM and Megatron Bridge.
Experience with FSDP, tensor, pipeline, context, or data parallelism; heterogeneous or per-module parallelism; optimizer research; or efficient sparse or long-context attention.
Experience with LLM/VLM post-training, including supervised fine-tuning, reinforcement learning for LLMs such as PPO, GRPO, or asynchronous RL, and large-scale RL frameworks such as NeMo-RL.
Experience with inference efficiency and model compression techniques, including quantization (FP8, NVFP4, INT4), pruning, knowledge distillation, neural architecture search, or diffusion and non-autoregressive language models.
Contributions to open-source AI frameworks such as Megatron-LM, Megatron Bridge, NeMo-RL, or Hugging Face Transformers.
Experience with GPU performance optimization, distributed systems, latency and throughput analysis, or profiling large-scale AI workloads.

Responsibilities

Design scalable systems for preparing high-quality multimodal datasets for foundation model training.
Develop algorithms and systems that improve the scalability, efficiency, and cost of large-scale pre-training and post-training.
Advance inference techniques to improve performance, reduce deployment cost, and support efficient serving across cloud and edge platforms.
Develop reusable infrastructure and contribute new model support to NVIDIA's open-source GenAI training platform.
Collaborate with research, product, and infrastructure teams to design algorithms, optimize systems, and contribute to NVIDIA's open-source AI stack.