NVIDIA seeks a Senior Software Engineer for AI Resiliency to lead development of AI software resiliency for powerful AI supercomputers.  Responsibilities include implementing and optimizing software features (checkpoint-recovery, error detection, isolation, straggler/hang detection) for 100,000+ GPUs, contributing to large-scale distributed systems with C++ and Python, working on AI system error handling, and collaborating with various teams.  The role involves developing and implementing robust tests, supporting production deployments, and ensuring seamless operation of AI training and inference workloads.  The ideal candidate possesses strong experience in distributed systems, parallel programming, and fault tolerance.

Explore gaming industy jobs in one of the leading Game Studios.

NVIDIA

Data Analyst, Internal Audit

Senior Network Engineer - Deployments

Manager, Systems Software

Senior Compiler Engineer - Deep Learning

Senior ASIC Design Engineer

Senior Infrastructure Software Engineer, Deep Learning Libraries

Senior High-Performance System Architect

Senior Software Engineer, Hardware Tools and Methodology Development

Senior Hardware Validation Engineer

Silicon Power Performance Engineer

<div data-automation-id="jobPostingDescription" class="css-4r17ng">We are now looking for a Senior Software Engineer for AI Resiliency.At NVIDIA, we are pushing the boundaries of what’s possible in AI. We are currently seeking a Senior Software Engineer to lead the development of AI software resiliency for the most powerful AI supercomputers in the world. As a member of our AI Software Resiliency team, you will play a pivotal role in defining and implementing critical resiliency features for AI supercomputers at a scale of 100,000+ GPUs. Your expertise will be crucial in driving down cluster downtime towards zero, ensuring that our AI systems remain robust and reliable at all times.What You’ll Be Doing:<ul><li>Develop AI Software Resiliency Features: Implement and optimize software features that improve AI system reliability at a massive scale, such as fast checkpoint-recovery, error detection, error isolation, and straggler/hang detection.</li><li>Hands-On Coding &amp; Optimization: Contribute to large-scale distributed systems with high-quality, production-level C++ and Python code. Enhance performance for AI workloads running on thousands of GPUs.</li><li>Fault Tolerance &amp; Debugging: Work on AI system error handling, implementing techniques to detect silent data corruption (SDC) and other failure scenarios. Assist in developing monitoring tools for proactive failure mitigation.</li><li>Collaborate Across Teams: Work closely with senior engineers, AI researchers, and hardware/software teams to integrate resiliency features into AI frameworks like PyTorch and JAX/XLA.</li><li>Testing &amp; Automation: Develop and implement tests to ensure robustness, scalability, and efficiency of resiliency mechanisms. Contribute to CI/CD pipelines to automate validation of AI workloads.</li><li>Support Production Deployments: Assist in debugging and performance tuning large-scale AI workloads in cloud and HPC environments, ensuring seamless operation of AI training and inference workloads.</li></ul>What We Need to See:<ul><li>You've achieved a Bachelor’s, Master’s or PhD in Computer Science, Electrical Engineering, or a related field, or equivalent experience.</li><li>Proficiency in C++ and Python, with experience in writing efficient, high-performance code.</li><li>6+ years of relevant experience</li><li>Strong understanding of distributed systems concepts, parallel programming, and fault tolerance in large-scale computing environments.</li><li>Familiarity with AI frameworks such as PyTorch, JAX/XLA, TensorFlow, or similar.</li><li>Experience with debugging and profiling tools (e.g., gdb, perf, valgrind, NVIDIA Nsight).</li><li>Excellent problem-solving skills and ability to work in a fast-paced, highly collaborative environment.</li></ul>Ways to Stand Out From the Crowd:<ul><li>Hands-on experience in training models or working with model training teams.</li><li>Hands-on experience with CUDA, NCCL, or MPI for GPU-accelerated computing, especially at extreme-scale.</li><li>Knowledge of checkpointing strategies, error mitigation, or fault-tolerant computing in AI training.</li><li>Experience working with large-scale AI clusters, HPC environments, or cloud-based AI workloads.</li><li>Strong systems programming skills and experience with low-level performance tuning.</li></ul>As part of the AI Resiliency team at NVIDIA, you’ll work alongside world-class engineers solving some of the hardest challenges in AI infrastructure. You’ll have the opportunity to contribute directly to making AI training and inference more reliable, scalable, and efficient. If you're passionate about AI, distributed systems, and high-performance computing, we want to hear from you!The base salary range is 184,000 USD - 287,500 USD. Your base salary will be determined based on your location, experience, and the pay of employees in similar positions.<div><div><div><div><div><div><div>You will also be eligible for equity and <a rel="noopener noreferrer" target="_blank" href="https://www.nvidia.com/en-us/benefits/">benefits</a>. NVIDIA accepts applications on an ongoing basis. </div></div></div></div></div></div></div>NVIDIA is committed to fostering a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate (including in our hiring and promotion practices) on the basis of race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law.</div>

Senior Software Engineer, AI Resiliency