CUDA Kernel Engineer for Massive GPU Compute

Pantera Capital; Palo Alto (CA); On-site; USD 180,000 - 440,000; Full time

Benefits offered by this job

Comprehensive medical, vision, and dental coverage

401(k) retirement plan

Equity options

Disability insurance

Life insurance

Job summary

Pantera Capital in Palo Alto is seeking a talented engineer to revolutionize AI supercomputing infrastructure. You’ll be tasked with designing, building, and optimizing massive GPU clusters that drive AI training and inference. The role demands hands-on expertise with CUDA kernels and requires collaboration with AI research teams to ensure top performance.

Offering a competitive salary between $180,000 - $440,000, additional benefits include comprehensive medical coverage, equity, and a 401(k) plan among others.

Qualifications

• Strong experience in systems programming.
• Proven involvement with large-scale GPU infrastructure.
• Hands-on experience in optimizing GPU kernels.

Responsibilities

• Design and optimize GPU clusters for AI workloads.
• Develop low-level CUDA kernels for optimal performance.
• Collaborate with research teams for scalability.

Skills

Deep low-level systems programming (C/C++/PTX/SASS)

Experience with large-scale GPU clusters or distributed compute infrastructure

GPU kernel optimization (CUTLASS, custom kernels)

Debugging and profiling skills using Nsight

Ability to optimize memory-bound and compute-bound scenarios

Job description

Pantera Capital in Palo Alto is seeking a talented engineer to revolutionize AI supercomputing infrastructure. You’ll be tasked with designing, building, and optimizing massive GPU clusters that drive AI training and inference. The role demands hands-on expertise with CUDA kernels and requires collaboration with AI research teams to ensure top performance.

Offering a competitive salary between $180,000 - $440,000, additional benefits include comprehensive medical coverage, equity, and a 401(k) plan among others.