AI Compiler Engineer

Engineer, AI Framework Software (ML Accelerator Compiler)

About the Role

We are looking for a motivated compiler engineer to join our AI compiler and toolchain team. You will work on modern compiler infrastructure for machine learning and heterogeneous accelerators, focusing on performance, correctness, and scalability of real production workloads.

This role is well suited for candidates with a strong foundation in C++ and algorithms, and an interest in compilers, systems, and performance engineering.

• Design, implement, and maintain compiler passes and transformations for AI and accelerator backends.
• Develop and optimize IR-level and backend-level optimizations (e.g., graph lowering, scheduling, fusion, memory planning).
• Work closely with hardware architects and runtime teams to enable new hardware features.
• Debug and resolve functional and performance issues in compiler pipelines.
• Contribute to tooling, testing infrastructure, and automation for compiler development.
• Analyze model workloads and identify optimization opportunities for real customer use cases.

• Master’s degree in Computer Science, Computer Engineering, or a related field (new graduates are welcome).
• Strong proficiency in C++.
• Solid background in data structures and algorithms
  .
• Good understanding of at least one of the following areas:
• compilers,
• operating systems,
• or high-performance systems.
• Ability to read, understand, and work with large codebases.
• Strong problem-solving skills and good engineering discipline.

• Research or project experience in compiler technologies (e.g., compiler courses, research projects, or open-source contributions).
• Hands‑on experience with compiler infrastructures or intermediate representations.
• Familiarity with machine learning frameworks or model formats.
• Experience with performance analysis and profiling tools.
• Exposure to parallel computing, GPUs, or custom accelerators.

• End‑to‑end AI compiler pipelines, from model frontends to hardware‑specific code generation.
• Performance tuning and memory optimization for modern ML workloads.
• How to build production‑quality compiler software used by real customers.
• Curious about how software maps onto hardware.
• Willing to learn new compiler and hardware concepts quickly.
• Enjoys collaborating across teams (compiler, runtime, hardware, and model teams).