AI Kernel Engineering Manager

This is a fully remote position, open to applicants in California.

📋 Description

• Lead, recruit, mentor, and develop a team of around 12 kernel engineers across NVIDIA, AMD, and ASIC hardware.

• Take responsibility for the MAX kernel roadmap and its successful delivery.

• Achieve or surpass the performance benchmarks set by open-source kernel libraries for each critical business model and hardware target.

• Support a portable kernel architecture that includes TileTensor, TensorEngine, TileIO, MegaFFN, and attention building blocks.

• Collaborate with Framework, Serve, Compiler/Runtime, Hardware Enablement, and Qualcomm kernel teams on priorities, interfaces, and escalations.

• Promote AI-assisted kernel development through kernel agents, fuzz verification, and playbooks.

• Foster the kernel engineering community within Modular and Qualcomm.

• Take charge of career development, performance management, and team wellbeing.

⛳️ Requirements

• Over 3 years of experience as an engineering manager leading teams focused on GPU kernel, compiler, HPC, or ML performance engineering.

• Excellent cross-functional communication skills with the ability to influence technical decisions alongside technical leads.

• Proficient in conveying status updates, risks, and tradeoffs to engineering leadership and customers.

• Proven history of hiring, retaining, and developing senior kernel or performance engineers.

• Hands-on experience in writing and optimizing GPU/accelerator kernels using CUDA, HIP/ROCm, Triton, CUTLASS/CuTe, or comparable technologies.

• Profound understanding of profiling, benchmarking, and roofline analysis.

• Candidates must reside in the United States.

• Capability to work remotely from home or from the Los Altos, CA office.

• Willingness to participate in in-person onboarding in Los Altos, CA.

• Readiness to travel 2–4 times annually.

• Beneficial but not mandatory: Experience with Mojo or other kernel and tile-level programming models.

• Beneficial but not mandatory: Multi-target experience beyond NVIDIA, including AMD, NPUs/ASICs, or edge/on-device.

• Beneficial but not mandatory: Involvement in AI-assisted or agentic kernel development.

• Beneficial but not mandatory: Contributions to open-source kernel libraries or inference engines.

🏝️ Benefits

• Comprehensive healthcare coverage.

• Retirement and savings programs.

• Opportunities for employee stock purchases.

• Paid time off.

• Resources for wellbeing.

• Programs to support families.

• Opportunities for learning and development.

• RSU grants/equity.

• Annual target bonus.

• Regular team onsite meetings and local gatherings.

• Travel 2–4 times a year.

• Remote work from home options.

• In-person onboarding in Los Altos, CA.