Inference Performance Engineer

Job Summary

Optimizing AI inference configurations, the full-time Inference Performance Engineer will enhance performance on large-scale benchmarks by developing reusable workflows and methodologies, profiling workloads, and collaborating with cross-functional teams, with opportunities for both onsite and remote work.

Key responsibilities

• Develop reusable skills and workflows for autonomous AI agents to optimize performance in AI inference workloads
• Measure and enhance throughput-per-GPU and user interactivity through various optimization techniques and configurations
• Collaborate with multiple teams to translate profiling insights into effective performance improvements across NVIDIA's GPU platforms

Required qualifications

• BS, MS, or PhD in Computer Science, Computer Engineering, Electrical Engineering, Applied Math, or a related field, or equivalent experience
• 3+ years of relevant engineering experience in AI model execution optimization
• Extensive knowledge of performance benchmarking and profiling GPU workloads using tools like Nsight Systems and PyTorch profiler
• Strong Python engineering skills and experience with large C++/CUDA codebases
• Rigorous experimental methodology for controlled comparisons and reproducible benchmarks