Deep Learning Compiler Engineer (New Grad)

Job Title

Deep Learning Compiler Engineer (New Grad)

Role Summary

As a new-graduate Deep Learning Compiler Engineer on Quadric's compiler team, you will develop and maintain compiler passes in CGC that lower ONNX/Relay IR to C++ targeting the Chimera GPNPU. Work includes IR transformations, code generation, debugging generated C++, and collaborating with hardware, kernel, and data science teams.

This is a hands-on engineering role on a small, senior team; changes ship to silicon and in-person collaboration from the Burlingame office is expected.

Experience Level

Entry-level / New graduate. Candidates should have completed or be completing a relevant degree within the past year or within six months.

Responsibilities

Primary responsibilities include designing and implementing compiler passes, debugging end-to-end compilation, and improving compiler decisions and toolchain quality.

• Design and implement IR transformations to lower neural network IR to GPNPU-targeted code and maintain pipeline components.
• Diagnose and debug compilation issues from generated C++ back through the pipeline using IR dumps, static analyses, and the ISS.
• Optimize data movement and core utilization in collaboration with senior engineers to improve generated code quality.
• Collaborate with kernel, hardware, and data science teams to align compiler features with model and hardware constraints.
• Improve test infrastructure, debugging utilities, and developer ergonomics across the compiler and runtime.

Requirements

Must-have technical skills and experience.

• Strong proficiency in Python and C++.
• Solid grasp of compiler concepts: intermediate representations, dataflow analysis, transformation passes, and lowering.
• Comfort reading and reasoning about large, unfamiliar codebases.
• Strong debugging and problem-solving skills; able to document and communicate findings clearly.
• Willingness to work on-site periodically; preference for candidates who can collaborate from the Burlingame office.

Nice-to-have:

• Coursework, research, or project experience in compilers, program analysis, or domain-specific languages.
• Experience with ML compiler frameworks such as TVM, MLIR, XLA, Glow, or IREE; experience writing non-trivial passes is a plus.
• Familiarity with neural network quantization, fixed-point arithmetic, or numerical analysis for ML.
• Experience with hardware-aware code generation for accelerators (GPU, DSP, NPU), assembly, or instruction scheduling.
• Previous internships in compilers, ML systems, or performance engineering; published research or open-source contributions.

Education Requirements

Bachelor's, Master's, or PhD in Computer Science, Electrical Engineering, or a related technical field. Degree should be completed within the past year or be completing within the next six months (new-graduate requirement).

About the Company

Company: Quadric

Headquarters: Burlingame, California, United States

Quadric is building the world’s first supercomputer designed for the real-time needs of edge devices. Founded in 2016, the company empowers developers across industries with innovative general-purpose neural processing unit (GPNPU) architecture for neural network workloads. Co-founded by technologists from MIT and Carnegie Mellon, Quadric aims to enable groundbreaking technology development.