CUDA Kernel Engineer

Location: Palo Alto Office, United States

Company

hirify.global develops world models and AI agents for understanding and building hardware, electronics systems, and semiconductors.

What you will do

• Develop, integrate, and optimize CUDA kernels for AI models used in semiconductor design and verification.
• Improve large-scale model training, inference, and reinforcement learning systems running across thousands of GPUs.
• Build performance tools, benchmarks, and integration layers to maximize GPU utilization.
• Develop GPU-accelerated primitives for graph reasoning, symbolic computation, and hardware simulation.
• Collaborate with AI researchers and semiconductor experts to translate domain-specific workloads into high-performance GPU code.
• Release kernels and tooling as contributions to open-source AI and HPC ecosystems.

Requirements

• Experience writing and optimizing CUDA kernels for large-scale AI workloads.
• Experience profiling and optimizing GPU performance for custom compute- or memory-bound workloads.
• Experience integrating custom kernels with training and inference frameworks such as PyTorch, Megatron, vLLM, or TorchTitan.
• Experience with current NVIDIA hardware and software stacks, including Hopper, Blackwell, NVLink, NCCL, or Triton.
• Ability to collaborate with AI researchers and semiconductor specialists on high-performance GPU implementations.

Culture & Benefits

• Work alongside researchers, engineers, and semiconductor experts from leading academic and technology organizations.
• Contribute to open-source AI and high-performance computing ecosystems.
• Work on AI systems designed to reason about circuit layouts, generate and validate RTL, and optimize chip architectures.