Quadric has created an innovative general purpose neural processing unit (GPNPU) architecture. Quadric's co-optimized software and hardware is targeted to run neural network (NN) inference workloads in a wide variety of edge and endpoint devices, ranging from battery operated smart-sensor systems to high-performance automotive or autonomous vehicle systems. Unlike other NPUs or neural network accelerators in the industry today that can only accelerate a portion of a machine learning graph, the Quadric GPNPU executes both NN graph code and conventional C++ DSP and control code.

The Role
As a Software Engineer Intern - Kernels, you will work closely with our senior AI Kernel Engineers to help enable a variety of AI/LLM models to run efficiently on the Quadric platform. This is a hands-on role where you will dive deep into hardware architecture and optimization techniques. You will gain invaluable experience developing, profiling, and optimizing kernel code, directly contributing to the performance of our AI inference stack. Note: Our preference is for a candidate willing to relocate to the California Bay Area who can regularly collaborate from our Burlingame office.

Responsibilities

• Develop & Implement: Assist in developing AI/LLM kernels and operators on the Quadric platform for efficient inference.
• Analyze & Profile: Help profile kernel performance across compute, data, and parallelism to identify micro-architecture and software bottlenecks.
• Code Optimization: Work alongside senior engineers to optimize C/C++ code to maximize hardware utilization for different workloads.
• Collaborate: Partner across related areas of the AI inference stack to support team priorities and business goals.
• Toolchain Contribution: Contribute to improvements in the Quadric toolchain, compiler, and runtime.