About the job

Flux Computing is seeking a Senior Performance Modelling Engineer to design and own analytical and simulation models that drive the evolution of its Optical Tensor Processing Unit (OTPU) architecture and software. In this role, you will create functional and high-fidelity, cycle-accurate simulators of the optical compute system, perform workload analysis, hunt for bottlenecks, and conduct large-scale design-space explorations to generate actionable insights that shape software, hardware, and optical roadmaps. You will collaborate closely with hardware, compiler, and ML framework teams to ensure models reflect reality and performance goals are met, while also developing Python/C++ tooling for trace parsing, statistical analysis, and visualisation, and integrating models into CI pipelines for continuous performance smoke testing. The ideal candidate will have 5+ years of experience building performance or power models for CPUs, GPUs, ASICs, or accelerators, strong coding ability in C++ and Python, familiarity with discrete-event or cycle-accurate simulators (gem5, SystemC, or in-house tools), and a solid understanding of computer architecture fundamentals including memory systems, interconnects, and queuing theory. Additional strengths include experience with ML workloads (PyTorch, TensorFlow, JAX), ability to read RTL or schematics, and strong data-visualisation and communication skills to turn complex simulation data into clear insights. This role offers a competitive salary with stock options, a unique location bonus of up to £24,000 for living within 20 minutes of Flux’s King’s Cross HQ in London, visa sponsorship with full relocation support, BUPA healthcare and dental coverage, high-spec equipment, chef-prepared meals, pension plans, lifestyle benefits, and a dynamic, fast-paced work environment where you will help shape the future of AI computing.