Deep Learning Engineer

NVIDIA is seeking a motivated Deep Learning engineer to integrate advanced communication technologies into AI stacks like PyTorch, vLLM, SGLang, TRT-LLM, and veRL.

You will work with the team that developed communication libraries, such as NCCL and NVSHMEM, for scaling Deep Learning applications. Your customers will have diverse multi-GPU needs, ranging from training on scales up to 100K GPUs to inference at microsecond latency.

Responsibilities:

• Integrate new communication libraries features in AI frameworks: from PoC to performance analysis to production.
• Perform deep analysis of AI workloads and frameworks to identify multi-GPU communication requirements and opportunities.
• Author custom communication or fused compute-communication kernels to showcase ultimate performance on NV platforms.
• Conduct in-depth research to achieve SOL GPU performance.
• Build fault-tolerant and elastic solutions for large-scale or dynamic AI workloads.
• Collaborate with a very dynamic team across multiple time zones.

Requirements:

• Pursuing a M.S. or Ph.D. in CE/CS/EE with a strong background in communication, kernel authoring, and/or AI training/inference.
• Rapid prototyping and development with Python, C++, CUDA or related DSLs (Triton, cuTe).
• Solid understanding of LLM models and parallelisms.
• Adaptability and passion to learn new areas and tools.
• Flexibility to work and communicate effectively.

Preferred Qualifications:

• Development experience with frameworks such as PyTorch, JAX, TRT-LLM, vLLM, SGLang, or veRL.
• Experience with DL communication patterns such as Expert Parallelism (EP), TP, DP & PP.
• Experience with CUDA kernel optimization and profiling.
• Experience with large-scale training or production inference stack.

NVIDIA offers highly competitive salaries and a comprehensive benefits package.