Senior AI Kernel Engineer - Edge Inference

quadric, Inc; Burlingame (CA); Hybrid; CAD 155,000 - 382,000; Full time

Benefits offered by this job

Medical, dental, and vision plans

401(k) retirement plan

Unlimited PTO

Company lunches and stocked kitchen

Parking/Caltrain passes

Downtown Burlingame office

Competitive equity

Health and wellbeing programs

Job summary

Quadric seeks an AI Kernel Engineer to enable a large number of AI kernels to run efficiently on its platform. The role involves developing a high-throughput kernel library for diverse models, analyzing and optimizing performance across hardware configurations, and collaborating across the AI inference stack.

The candidate should have 5+ years in kernel development, solid C/C++ and Python skills, and experience with CUDA/DSP/NEON/Triton.

Qualifications

• Bachelor's or Master's in Computer Science and/or Electric Engineering.
• 5+ years of AI kernel development and optimization experience.
• Experience with model/kernel inference performance profiling.
• Experience with CUDA, DSP, NEON, or Triton-lang.
• Proficiency in C/C++ and Python; assembly knowledge a plus.
• Strong problem solving, debugging and communication skills.

Responsibilities

• Develop AI/LLM kernels/operators on Quadric platform for efficient inference.
• Optimize kernel performance for different hardware configurations and workloads.
• Profile and analyze kernel performance in terms of compute, data and parallelism; identify bottlenecks and provide optimization solutions.
• Optimize kernel C/C++ code to maximize hardware utilization.
• Collaborate across AI inference stack to support team and business priorities.
• Improve Quadric toolchain, compiler and runtime.
• Provide technical support and documents to customers and developer community.

Skills

C/C++

Python

Problem solving

Communication

Education

Bachelor's/Master's in CS or EE

Tools

CUDA

DSP

NEON

Triton-lang

Job description

Quadric seeks an AI Kernel Engineer to enable a large number of AI kernels to run efficiently on its platform. The role involves developing a high-throughput kernel library for diverse models, analyzing and optimizing performance across hardware configurations, and collaborating across the AI inference stack.

The candidate should have 5+ years in kernel development, solid C/C++ and Python skills, and experience with CUDA/DSP/NEON/Triton.