Bare-Metal Linux Performance Engineer (Low-Latency)

Miller Maxwell Ltd; England; On-site; GBP 90,000 - 130,000; Full time

Job summary

Miller Maxwell Ltd in the United Kingdom is seeking a Linux performance engineer to join a world‑class HFT infrastructure team. You will optimise bare‑metal Linux estates for deterministic, low latency across CPU, memory, and networking.

You’ll collaborate with quant researchers and C++ engineers, using perf, ftrace, eBPF, and flame graphs to drive microsecond improvements and predictable execution under pressure. This role offers access to cutting‑edge hardware and bespoke kernel builds.

Qualifications

• Mastery of Linux internals — you understand why the kernel behaves the way it does.
• Proven experience tuning systems for low‑latency or high‑throughput workloads.
• Strong networking fundamentals (TCP internals, multicast, kernel bypass).
• Ability to reason about hardware/software interactions at a microscopic level.

Responsibilities

• Deep kernel tuning: CPU isolation, IRQ affinity, scheduler behaviour, NUMA optimisation, hugepages, and custom kernel builds.
• High‑performance networking: DPDK, RDMA, kernel‑bypass stacks, NIC firmware tuning, and microburst mitigation.
• Hardware‑aware engineering: PCIe topology, cache coherency, memory bandwidth profiling, and BIOS‑level optimisation.
• Observability at nanosecond resolution: perf, ftrace, eBPF, flame graphs, and bespoke latency instrumentation.
• Collaborate with quant researchers and C++ engineers to reduce jitter and tail latency.

Skills

Linux kernel internals

Low-latency tuning

Networking fundamentals

Performance measurement

C/C++

Education

Bachelor's degree in Computer Science or related field

Tools

Perf

Job description

Miller Maxwell Ltd in the United Kingdom is seeking a Linux performance engineer to join a world‑class HFT infrastructure team. You will optimise bare‑metal Linux estates for deterministic, low latency across CPU, memory, and networking.

You’ll collaborate with quant researchers and C++ engineers, using perf, ftrace, eBPF, and flame graphs to drive microsecond improvements and predictable execution under pressure. This role offers access to cutting‑edge hardware and bespoke kernel builds.