AI Compiler Tech Lead

Darwin Recruitment; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Darwin Recruitment is seeking a hands-on Technical Lead to guide AI compiler development for an AI accelerators project in San Jose, California.

You will lead technical delivery across the compiler stack and stay closely involved in design and code. You will develop compiler passes, optimise scheduling and memory movement, and collaborate with hardware, architecture and kernel engineers to maximise performance.

Qualifications

• Strong experience developing compilers for AI accelerators, NPUs, GPUs or comparable hardware.
• Strong C++ skills, with Python experience beneficial.
• Understanding of compiler intermediate representations, optimisation passes and code generation.
• Familiarity with frameworks or toolchains such as LLVM, MLIR, TVM or equivalent technologies.
• Experience with neural network workloads, quantisation and accelerator memory constraints.
• Experience leading technical projects and supporting other engineers.
• Ability to connect compiler design decisions with hardware performance.

Responsibilities

• Lead technical delivery across the AI compiler stack, remaining hands‑on with design and development.
• Develop compiler passes for graph optimisation, operator lowering and hardware‑specific code generation.
• Improve model execution through efficient scheduling, memory allocation and data movement.
• Work closely with hardware, architecture and kernel engineers to optimise performance.
• Support model compatibility, numerical accuracy and debugging across the compilation pipeline.
• Mentor engineers, review code and guide technical decisions.

Skills

AI compilers

C++

Python

Compiler IRs

Code generation

LLVM/MLIR/TVM

Neural networks

Performance optimization

Leadership

Tools

LLVM

MLIR

TVM

Job description

Technical Lead, AI Compiler

AI Accelerators | San Jose, California

We’re supporting a semiconductor company developing AI processing technology. They’re looking for a hands‑on Technical Lead to guide AI compiler development and help machine learning workloads run efficiently on dedicated accelerator hardware.

Responsibilities

• Lead technical delivery across the AI compiler stack, remaining hands‑on with design and development.

• Develop compiler passes for graph optimisation, operator lowering and hardware‑specific code generation.

• Improve model execution through efficient scheduling, memory allocation and data movement.

• Work closely with hardware, architecture and kernel engineers to optimise performance.

• Support model compatibility, numerical accuracy and debugging across the compilation pipeline.

• Mentor engineers, review code and guide technical decisions.

Requirements

• Strong experience developing compilers for AI accelerators, NPUs, GPUs or comparable hardware.

• Strong C++ skills, with Python experience beneficial.

• Understanding of compiler intermediate representations, optimisation passes and code generation.

• Familiarity with frameworks or toolchains such as LLVM, MLIR, TVM or equivalent technologies.

• Experience with neural network workloads, quantisation and accelerator memory constraints.

• Experience leading technical projects and supporting other engineers.

• Ability to connect compiler design decisions with hardware performance.

The Opportunity

• Take a technical leadership role in AI compiler development.

• Influence how machine learning models map onto accelerator hardware.

• Work closely with teams across software and silicon development.