Location
Central London

Hours
Full Time

Salary
Competitive, commensurate with experience

About the Role
Fuse Energy is a forward-thinking renewable energy startup on a mission to deliver a terawatt of renewable energy - fast. Combining first-principles thinking with cutting-edge technology, we are building a radically better energy system. With $210M raised from top-tier investors, we are expanding into high-performance compute infrastructure at the intersection of energy and AI. This role focuses on optimising power-dense GPU workloads by writing and tuning low-level CUDA code that powers our inference workloads.

As a CUDA Engineer, you will design custom CUDA kernels, tune performance across memory bandwidth and compute bottlenecks, and maximise throughput on every GPU in our fleet. You will work at the level of streaming multiprocessors (SMs), warps, and memory hierarchies to deliver high-performance solutions that enable Fuse to scale its compute infrastructure rapidly and reliably.

The opportunity lies in addressing the growing demand for high-performance compute capacity, where speed, power efficiency, and reliability are critical. Your work will be central to how Fuse scales its GPU compute infrastructure to meet market needs.

Responsibilities
- Write and optimise custom CUDA kernels for core transformer inference operations
- Profile kernels to identify and eliminate bottlenecks in occupancy, memory throughput, and warp divergence
- Apply kernel fusion to reduce memory round-trips and launch overhead across inference pipelines
- Optimise memory access patterns and manage the memory hierarchy for maximum bandwidth utilisation
- Implement quantisation-aware kernels and mixed-precision arithmetic to reduce latency and memory footprint
- Build and tune caching mechanisms for efficient autoregressive decoding
- Tune kernel launch configurations for target GPU architectures
- Benchmark kernels against existing baselines and drive measurable throughput and latency improvements
- Write tests for CUDA code to catch performance and correctness regressions
- Maintain internal CUDA libraries and contribute to team coding standards and documentation

Experience
- 4+ years writing production CUDA code with a proven track record of shipping performance-critical kernels
- Deep understanding of GPU microarchitecture, warps, occupancy, register pressure, and memory hierarchy
- Strong CUDA C++ skills including streams and asynchronous execution
- Hands-on experience profiling to diagnose compute-bound vs. memory-bound bottlenecks
- Experience with kernel fusion, memory coalescing, and avoiding warp divergence
- Experience writing quantised and mixed-precision kernels
- Solid grasp of parallel algorithm design and numerical precision tradeoffs

About you
- Passionate about high-performance GPU programming and optimisation
- Detail-oriented with strong problem-solving skills
- Comfortable working in a fast-paced startup environment
- Collaborative team player who contributes to coding standards and documentation

Qualifications
- Nice to have:
- Experience with transformer/attention-style kernels or autoregressive decoding
- Experience building high-performance GPU libraries from scratch
- Background in HPC or other latency-critical performance engineering
- Exposure to multi-GPU or multi-node kernel-level optimisation
- Comfortable reading PTX/SASS to validate kernel efficiency