Senior ML Accelerator Engineer - GPU
About The Position
The AI Kernels & Compilers team helps turn perception, prediction, and planning research into production-grade software that runs efficiently and reliably on vehicles. The AI Kernels team develops high-performance GPU kernels and custom libraries for on-vehicle ML inference in ADAS and autonomous driving. The work focuses on improving core AI workloads while meeting real-world latency, throughput, and reliability requirements, and integrating kernels into the ML runtime stack and broader AV software stack. This role partners with AI Solutions, AI Compilers, AI Architecture, AI Tooling, and other teams to deliver reusable, reliable, high-performance libraries for production vehicles. The position is categorized as hybrid. The selected candidate is expected to report to a specific location at least three times per week, or at another frequency directed by their manager. Travel is required less than 25% of the time. Relocation benefits are available to candidates who qualify under company policy.
Requirements
- At least 2 years of relevant industry experience or equivalent experience.
- BS, MS, or PhD in computer science or a related technical field.
- Excellent GPU programming skills in CUDA and a thorough understanding of parallel programming patterns and GPU architecture.
- Hands-on experience benchmarking, profiling, debugging, and optimizing accelerator libraries and kernels using the NSight suite or similar tools.
- Strong background in software architecture, library design, and design patterns.
- Strong C++ programming skills and ability to work comfortably in large codebases.
- Solid background in system performance, high-performance computing, and/or architecture-aware optimizations.
- Strong communication skills and ability to work collaboratively within a team.
- Excellent analytical and problem-solving skills.
Nice To Haves
- At least 2 years of relevant industry experience or equivalent experience.
- Experience with tensor core programming, CUTLASS, and/or CuTe.
- Experience with ML model architectures, particularly transformer-based models.
- Experience with low-latency or real-time systems.
- Experience with lower levels of an accelerator software stack, such as drivers, runtimes, and compilers.
Responsibilities
- Design, implement, benchmark, and iterate on CUDA-based kernels and custom operators to improve performance for on-vehicle inference workloads.
- Build and improve tooling and infrastructure for profiling, debugging, and validating CUDA kernels and accelerator-backend code across the AV stack.
- Partner with AI Solutions, Compilers, and Architecture to translate model and system requirements into kernel roadmaps, priorities, and project plans.
- Collaborate with compiler, performance tooling, runtime, and deployment solutions teams to deliver reusable, reliable, high-performance libraries into production.
- Maintain technology standards, methodologies, processes, and guidelines for GPU kernel development and performance engineering through code review.
- Manage relationships with internal customers to ensure kernels and libraries meet real-world needs.
Benefits
- Medical, dental, and vision benefits.
- Health Savings Account and Flexible Spending Accounts.
- Retirement savings plan.
- Sickness and accident benefits.
- Life insurance.
- Paid vacation and holidays.
- Tuition assistance programs.
- Employee assistance program.
- GM vehicle discounts.
- Incentive pay program with payouts based on company performance, job level, and individual performance.
- Relocation benefits for candidates who qualify under company policy.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
