C++ Engineer (GPU and Compute Architect focus) - HFT

On behalf of a leading HFT firm, we are supporting the formation of a specialised performance engineering team focused on optimising the most compute-intensive R&D workloads. This role sits at the intersection of systems architecture, software engineering, and quantitative research, with a mandate to push the boundaries of throughput, latency, and resource efficiency across heterogeneous compute environments.

This is an ideal opportunity for engineers with a background in GPU architecture, HPC systems, or semiconductor performance tuning to apply their expertise in one of the most latency-sensitive domains in tech.

Core Responsibilities

• Architect and optimise distributed compute pipelines across CPU/GPU, memory, storage, and interconnects
• Profile and tune performance of AI/ML, simulation, and research workloads across containerised environments
• Collaborate with Software and Infra teams to identify architectural bottlenecks and implement low-level improvements
• Evaluate and integrate HPC technologies to support scale-out and latency-sensitive workloads

• 3 - 10 years of professional proficiency in Python and C++
• Hands-on experience with GPU compute
• Familiarity with HPC storage stacks and RDMA/networking frameworks
• Background in semiconductor performance engineering (e.g., microarchitecture-level tuning, driver/kernel optimisation)

If you've worked on optimising GPU compute/HPC systems and want to apply that expertise in one of the most demanding and intellectually rigorous environments in finance, please apply.