Summary
✨ AI‑Generated
An experienced C++ and Linux performance engineer is sought to optimize ultra-low-latency systems where performance is measured at the nanosecond level. The role covers Linux kernel and scheduler optimization, CPU/cache/NUMA analysis, network performance, low-level profiling, kernel tuning, and cross-stack optimization with hardware specialists.
Highlights
Work on highly performance-sensitive trading infrastructure at the intersection of modern C++, Linux internals, CPU architecture, networking, and hardware, with deep exposure to low-level optimization and profiling.
Description
A world-class quantitative trading firm is seeking exceptional C++ and Linux performance engineers to work on the infrastructure underpinning some of the fastest trading systems in the industry.
This is an environment where performance is measured in nanoseconds, not milliseconds.
Every cache miss, scheduler decision, memory allocation, kernel interrupt, and network hop is scrutinised.
You'll work at the intersection of C++, Linux internals, CPU architecture, and hardware, extracting deterministic performance from modern x86 systems.
What You'll Own
Ultra-low-latency C++ trading and market data infrastructureLinux kernel, scheduler, and network stack optimisationCPU, cache, NUMA, and memory performance analysisLow-level profiling using perf, eBPF, ftrace, and hardware countersKernel tuning, CPU isolation, IRQ affinity, and latency reductionCross-stack optimisation alongside FPGA, hardware, and infrastructure specialistsWhat They're Looking For
3-12 years of modern C++ (C++17/20)Strong Linux systems programming expertiseDeep understanding of Linux internals and performance engineeringExperience debugging bottlenecks across kernel, CPU, memory, and hardware layersKnowledge of CPU architecture, networking, and low-latency systemsElite academic credentials: First-Class degree (or equivalent GPA) from a top university in Computer Science, Mathematics, Physics, Engineering, or a related STEM field
Why Join?
This role is about understanding how machines actually behave under load and removing latency at every layer of the stack.
You'll profile production systems, tune kernels, analyse CPU behaviour, and optimise performance where microseconds are already too slow.
Apply to learn more.
