About the job

Flux Computing designs and manufactures optical processors to train and run inference on large AI models, and is seeking Staff Compiler Engineers to architect and build the ML backend—including compiler, runtime, and debugger—for its next-generation OPTUs. In this role, you will own integration with PyTorch, TensorFlow, JAX, and MXNet down to low-level kernel drivers, creating seamless support for a broad ecosystem of large AI models while partnering with hardware and modelling teams to optimize performance. Key responsibilities include defining and delivering the technical vision and roadmap, collaborating with hardware engineers to align compiler and runtime requirements with device design, designing and implementing compiler tools for multiple frameworks, applying advanced optimisation techniques such as layout, fusion, scheduling, and tiling, mentoring peers and establishing engineering standards, and guiding the direction of technology by staying ahead of GPU, AI accelerator, and optical computing trends. The ideal candidate will have 7+ years of experience in software engineering with expertise in C/C++, ML framework internals, compilers, and low-level optimisation techniques, along with hands-on experience with toolchains like LLVM, OpenXLA/XLA, MLIR, and TVM. Strong problem-solving ability, teamwork, and communication skills are essential, as well as a degree in computer science, electrical engineering, telecoms engineering, mathematics, or a related field; personal projects are highly valued. Flux offers a competitive salary and stock options, relocation and visa sponsorship, healthcare and dental cover, high-spec equipment, chef-cooked meals, social events, additional compensation for living close to the office, and many other benefits designed to support productivity, comfort, and work-life balance.