Senior AI Compiler Engineer - Equity & ML Backend

Flux Computing; Greater London; On-site; GBP 90,000 - 140,000; Full time

Benefits offered by this job

Equity options

Living-Local Bonus

Retirement benefits

Job summary

OLIX is seeking a Senior Compiler Engineer to join the team building the ML backend (compiler, run-time, debugger) for our next‑gen hardware. You will enable seamless PyTorch, TensorFlow, JAX integration, and drive performance optimisations on bespoke accelerators.

You will lead design and implementation of the compiler stack, collaborate with hardware and modelling teams, and mentor peers to maintain coding standards as the company scales its decode accelerator DX‑1 project.

Qualifications

• 5+ years of software engineering focusing on C/C++.
• Deep expertise in ML framework internals, compilers and optimisation.
• Strong experience with TensorFlow, PyTorch or JAX.
• Hands-on with LLVM, MLIR, OpenXLA/XLA, TVM toolchains.
• Proven ability to optimize large AI models in HPC contexts.
• Excellent problem-solving and communication skills.
• Demonstrated teamwork in fast-paced environments.
• Bachelor's degree in CS/EE or related field; strong personal projects.

Responsibilities

• Own and deliver compiler, runtime and debugger for ML backend.
• Design and implement support for PyTorch, TensorFlow, JAX on custom hardware.
• Apply advanced techniques to improve throughput and efficiency.
• Lead code reviews and mentor peers on performance engineering.
• Stay ahead of GPU accelerator trends and prototype innovations.

Skills

C/C++ programming

ML framework internals

TensorFlow / PyTorch / JAX

LLVM / MLIR / XLA

High-performance computing

Problem solving

Team collaboration

Personal projects

Education

Bachelor's degree in computer science or related field

Tools

LLVM

OpenXLA/XLA

MLIR

TVM

Job description

OLIX is seeking a Senior Compiler Engineer to join the team building the ML backend (compiler, run-time, debugger) for our next‑gen hardware. You will enable seamless PyTorch, TensorFlow, JAX integration, and drive performance optimisations on bespoke accelerators.

You will lead design and implementation of the compiler stack, collaborate with hardware and modelling teams, and mentor peers to maintain coding standards as the company scales its decode accelerator DX‑1 project.