About this role
Salary: £63,000 - 103,000 per year
Requirements: 4+ years writing production CUDA code, with a track record of shipping performance-critical kernelsDeep understanding of GPU microarchitecture, warps, occupancy, register pressure, and memory hierarchyStrong CUDA C++ skills, including streams and asynchronous executionHands-on experience profiling to diagnose compute-bound vs. memory-bound bottlenecksExperience with kernel fusion, memory coalescing, and avoiding warp divergenceExperience writing quantised and mixed-precision kernelsSolid grasp of parallel algorithm design and numerical precision tradeoffsNice to have: experience with transformer/attention-style kernels or autoregressive decodingNice to have: experience building high-performance GPU libraries from scratchNice to have: background in HPC or other latency-critical performance engineeringNice to have: exposure to multi-GPU or multi-node kernel-level optimisationNice to have: comfortable reading PTX/SASS to validate kernel efficiency Responsibilities: Write and optimise custom CUDA kernels for core transformer inference operationsProfile kernels to identify and eliminate bottlenecks in occupancy, memory throughput, and warp divergenceApply kernel fusion to reduce memory round-trips and launch overhead across inference pipelinesOptimise memory access patterns and manage the memory hierarchy for maximum bandwidth utilisationImplement quantisation-aware kernels and mixed-precision arithmetic to reduce latency and memory footprintBuild and tune caching mechanisms for efficient autoregressive decodingTune kernel launch configurations for target GPU architecturesBenchmark kernels against existing baselines and drive measurable throughput and latency improvementsWrite tests for CUDA code to catch performance and correctness regressionsMaintain internal CUDA libraries and contribute to our team coding standards and documentation Technologies: AICUDASASSNodeJS More:
We are Fuse Energy, a forward-thinking renewable energy startup on a mission to deliver a terawatt of renewable energy fast. We combine first-principles thinking with cutting-edge technology to build a radically better energy system. Backed by $210M from top-tier investors including Multicoin, Balderton, Lakestar, Accel, Creandum, Lowercarbon, Ribbit, Box Group and strategic angels, we are expanding into high-performance compute infrastructure at the intersection of energy and AI. We focus on optimising how power-dense GPU workloads are scheduled, cooled, and balanced against grid conditions in real time. This is a full-time role with competitive salary, an equity sign-on bonus, a biannual bonus scheme, fully expensed tech to match your needs, and a breakfast and dinner allowance for office-based employees.
last updated 36 week of 2026