CareerPlan

AI Kernel Engineer Intern

Burlingame, CA (HQ)Internship$104k–$119k2026-10-23 → 2026-11-20
On-site

Core

Develop and optimize AI/LLM kernels and operators for efficient inference on a General Purpose Neural Processing Unit (GPNPU) architecture.

Role type

AI Kernel Engineer Intern

Builds

AI inference stack for edge devices (automotive, industrial, robotics)

Domain

Edge AI / Semiconductor / Hardware-Software Co-design

Deliverable

production ML models

Required skills

C/C++, Python, Computer Architecture, Hardware-Software Interaction, Kernel Development, Performance Profiling, Code Optimization

Preferred skills

CUDA, Triton-lang, Assembly Language, Model Inference Profiling, Hardware-oriented Software Development

Technologies

GPNPU, C++, Python, CUDA, Triton-lang

Responsibilities

Develop AI/LLM kernels and operators; Profile kernel performance across compute, data, and parallelism; Optimize C/C++ code for hardware utilization; Contribute to compiler and runtime improvements.

Seniority

Intern