Nvidia Principal Engineer AI Software Resiliency 

United States, Texas 

411515960

What You'll Be Doing:

• AI Resiliency Expertise: Serve as a trusted authority on AI software resiliency, guiding the architecture, modeling, and scoping of resiliency features to support frontier model training at scale.

• Feature Development: Lead the execution and development of software resiliency features, including fast checkpoint-recovery, automatic error detection, error isolation, SDC detection and mitigation, and straggler/hang detection.

• Software Engineering Leadership: Drive engineering excellence by contributing to large software codebases, ensuring high code quality, rigorous testing, and solving complex challenges. Lead others by example, fostering a culture of collaboration, innovation, and continuous improvement.

• Cross-Team Collaboration: Work closely with multiple teams and stakeholders across NVIDIA to align on mission requirements, provide regular updates, and ensure the successful integration of resiliency features into AI frameworks like PyTorch and JAX/XLA.

• Customer Engagement: Collaborate directly with major customers to embed AI resilience features into their AI frameworks, ensuring seamless integration and optimal performance.

• Product Delivery: Partner effectively with TPMs, PMs, and QA teams to ensure the timely and successful launch of resiliency features.

What We Need to See:

• Master’s or Ph.D. in Computer Science, Electrical Engineering, Computer Engineering, or a related field from a reputed institution, or equivalent experience.

• A minimum of 10 years of experience in systems architecture or related fields, with a deep understanding of distributed systems and large-scale AI infrastructure.

• At least 10 years of hands-on experience in software development for distributed systems and 5 years in developing AI frameworks such as PyTorch or JAX/XLA.

• Proven track record of working effectively across multiple engineering fields and communicate complex technical concepts to a diverse set of collaborators.

Ways to Stand Out From the Crowd:

• AI Supercomputing Expertise: Experience with large-scale AI supercomputing applications, including in-depth knowledge of AI workload training and inference requirements.

• System Architecture Passion: A strong passion for developing AI-specific system architectures, including CPUs, GPUs, memory, storage, and networking.

• Lifecycle Experience: Hands-on involvement in the design, development, and deployment of large-scale AI supercomputers.

• HPC Best Practices: Practical experience in adopting and implementing high-performance computing (HPC) software development in large-scale environments.