Software Engineer(NVIDIA)

Job Description

Bring up, validate, benchmark, and debug large-scale distributed AI training and inference workloads across NVIDIA GPU clusters. Develop and maintain benchmarking, automation, resilience, and failure-attribution tools, and use profiling and cluster analysis to recommend performance improvements.

Requirements: A bachelor's or master's degree in computer science or a related technical field, or equivalent experience, and at least three years of software development experience in AI, HPC, or systems-level applications are required. Candidates need strong Python and C/C++ skills, hands-on experience with multi-GPU or multi-node workloads, and experience debugging distributed systems and AI applications across the software and hardware stack.

Key Skills: Python, C/C++, Distributed Systems Debugging, Multi-GPU Computing, Multi-Node Workloads, CUDA, NCCL, PyTorch, NeMo, Megatron, TensorRT-LLM, Benchmarking, Root-Cause Analysis, Performance Optimization, RDMA, InfiniBand

Benefits: Equity, Benefits