Senior CUDA Kernel & ML Inference Engineer

General Motors; Austin (TX); Hybrid; USD 170,000 - 258,000; Full time

Benefits offered by this job

Health benefits

Retirement savings plan

Vehicle discounts

Tuition assistance

Employee assistance program

Relocation benefits

Paid vacation & holidays

Job summary

General Motors is seeking an experienced GPU/CUDA software engineer to advance on-vehicle AI kernels and libraries. You will design, optimize, and validate CUDA-based kernels, collaborating across compiler, runtime, and deployment teams to push performance on real vehicles.

The role emphasizes cross-functional teamwork, state-of-the-art kernel development, and ensuring latency and throughput targets are met in production environments.

Qualifications

• Minimum 2+ years of relevant industry experience or equivalent experience.
• BS, MS or PhD in CS, or related technical field.
• Excellent GPU programming skills in CUDA, with a thorough understanding of parallel programming patterns and GPU architecture.
• Hands-on experience benchmarking, profiling, debugging and optimizing accelerator libraries and kernels using NSight tools or similar.
• Strong background in software architecture, library design, and design patterns.
• Strong C++ programming skills with the ability to work in large codebases.
• Solid background in system performance and HPC/architecture-aware optimizations.
• Strong communication skills and the ability to work collaboratively within a team.

Responsibilities

• Design, implement, benchmark, and iterate on CUDA-based kernels and custom operators to maximize on-vehicle inference performance.
• Build and improve tooling to profile, debug, and validate CUDA kernels and accelerator-backend code across the AV stack.
• Collaborate with AI Solutions, Compilers, and Architecture to translate requirements into kernel roadmaps.
• Deliver reusable, reliable, high-performance libraries into production with cross-functional teams.
• Maintain high tech standards and code reviews for GPU kernel development and performance engineering.
• Manage relationships with internal customers to ensure kernels meet real-world needs.

Skills

CUDA programming

GPU kernel dev

C++ programming

Performance tuning

Parallel programming

Strong communication

Team collaboration

Education

BS in CS

MS in CS

PhD in CS

Tools

NSight tools

CUTLASS

CuTe

CUDA toolkit

Job description

General Motors is seeking an experienced GPU/CUDA software engineer to advance on-vehicle AI kernels and libraries. You will design, optimize, and validate CUDA-based kernels, collaborating across compiler, runtime, and deployment teams to push performance on real vehicles.

The role emphasizes cross-functional teamwork, state-of-the-art kernel development, and ensuring latency and throughput targets are met in production environments.