New-Grad Deep Learning Compiler Engineer (Edge AI)

SpeedyApply LLC; Burlingame (CA); On-site; USD 120,000 - 160,000; Full time

Benefits offered by this job

Competitive salary and equity

Healthcare plan starting day one

401(k) retirement plan

Unlimited PTO

Office in Burlingame near Caltrain

Monthly parking or Caltrain passes

Job summary

Quadric in Burlingame, CA is seeking a new-grad Deep Learning Compiler Engineer to advance CGC, Quadric's neural network compiler that lowers ONNX models through Relay IR to C++ for the Chimera GPNPU. You will own real compiler passes, design IR transformations, and ensure efficient mapping of neural networks to hardware.

This is a hands‑on role on a small, senior team with a fast feedback loop to silicon. Collaboration with kernel, hardware, and data science teams is essential, and the role

Qualifications

• Bachelor's, Master's, or PhD in CS/EE and recent completion.
• Strong Python and C++ skills.
• Solid compiler concept knowledge.
• Ability to read large, unfamiliar codebases.
• Excellent debugging and problem-solving skills.

Responsibilities

• Own compiler passes. Design and implement IR transformations that lower neural network IR to GPNPU‑targeted code.
• Debug end‑to‑end. Diagnose compilation issues by tracing problems from generated C++ back through the pipeline.
• Improve compiler decisions. Reduce data movement, improve core utilization, and tighten the gap between hardware capabilities and emitted code.
• Collaborate across teams. Partner with kernel, hardware, and data science teams to align features with model requirements.
• Strengthen the toolchain. Contribute to tests, debugging utilities, and developer ergonomics.

Skills

Python

C++

Compiler concepts

IR transformations

Debugging

Education

Bachelor's/Master's/PhD in CS/EE

Tools

MLIR

TVM

XLA

IREE

Job description

Quadric in Burlingame, CA is seeking a new-grad Deep Learning Compiler Engineer to advance CGC, Quadric's neural network compiler that lowers ONNX models through Relay IR to C++ for the Chimera GPNPU. You will own real compiler passes, design IR transformations, and ensure efficient mapping of neural networks to hardware.

This is a hands‑on role on a small, senior team with a fast feedback loop to silicon. Collaboration with kernel, hardware, and data science teams is essential, and the role