AI Inference Optimization Engineer
About The Position
As a member of Quadric’s AI Application team, the AI Inference Optimization Engineer gets AI models running quickly and accurately on Quadric’s General Purpose Neural Processing Unit (GPNPU). The work spans CNN perception models, LLMs, VLMs, and VLAs for automotive, robotics, and AI-PC applications. The role involves profiling real workloads, writing and tuning kernels, selecting quantization strategies, and developing repeatable model deployment tools and workflows. The role is based at Quadric’s downtown Burlingame, California office. Success in the first 6–12 months includes independently deploying five new models on the GPNPU while meeting accuracy targets, delivering a 2x speedup on a model through kernel, model compression, or dispatch work, and shipping a deployment or optimization tool used by the team every day. The base salary range is $160,000 to $195,000. The position is also eligible for equity and a discretionary annual performance bonus, as applicable to the role and level. Quadric describes its culture as collaborative and low-ego, with significant ownership and impact, and focused on innovative disruption.
Requirements
- At least 3 years of experience in AI model deployment or kernel optimization.
- BS/MS in computer science, computer engineering, or a related field.
- Strong C/C++ and Python skills.
- Hands-on performance analysis experience, including profilers, roofline and bandwidth reasoning, and identifying compute versus memory bottlenecks.
- Hands-on quantization experience, including PTQ/QAT and int8/int4, and debugging accuracy loss.
- Experience writing performance-critical kernels using CUDA, Triton, OpenCL, SIMD intrinsics, or DSP.
- Fluency with AI coding agents such as Claude Code or Cursor, and interest in building agent-driven workflows for kernel and model bring-up.
Nice To Haves
- Hands-on experience in pruning and sparsification.
- Solid understanding of accelerator microarchitecture, including memory hierarchy and DMA.
- Solid understanding of fixed-point and numerical methods, including mixed precision.
- Understanding of compiler code generation and kernel scheduling.
Responsibilities
- Analyze and optimize performance on the GPNPU.
- Deploy and profile GPNPU models, including vision and perception models, LLMs, VLMs, and VLAs.
- Analyze model accuracy and choose appropriate quantization schemes.
- Write and tune kernels and optimize model compute graphs on the GPNPU, balancing speed and accuracy.
- Create efficient workflows to convert and deploy models on the GPNPU.
- Develop tools and workflows to accelerate model deployment, including profiling, tracing, visualization, bottleneck identification, and agentic automation.
- Customize open-source model optimization and quantization tools.
- Integrate and automate model pruning, sparsification, and quantization in the workflow.
Benefits
- Medical, dental, and vision insurance from day one; premiums covered at 99% for employees.
- Company-paid life insurance.
- Voluntary supplemental life insurance.
- Short-term and long-term disability insurance.
- Commuter support, including parking or Caltrain reimbursement.
- Flexible spending account (FSA) and health savings account (HSA).
- Equity with the business.
- Paid parental leave.
- 401(k) retirement plan.
- Flexible PTO.
- Winter holiday shutdown.
- Catered lunch each day in the office.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Mid Level
