Low-Latency HFT Linux Engineer

DeepFin Research is a machine learning driven proprietary trading firm deploying cutting edge deep learning models across the global financial markets, combining deep learning signals with HFT execution and monetisation strategies. We have expanded rapidly across markets, products, teams, and jurisdictions, and are looking for top people to join our team.

We are seeking a Senior Linux Engineer to join our Infrastructure team. In this role you will own the full lifecycle of our bare-metal and co-located trading servers—from provisioning and hardening through to ultra-low-latency kernel tuning and real-time performance analysis. You will work at the intersection of DevOps, systems engineering, and network infrastructure, ensuring our trading systems operate with deterministic, sub-microsecond jitter in some of the most demanding production environments in the industry.

Infrastructure & Server Management

• Design, deploy, and maintain bare-metal Linux servers across co-location facilities with strict uptime and latency SLAs.
• Manage server provisioning pipelines using PXE boot, Kickstart/Preseed, and configuration management tools (Ansible, Salt, or Puppet).
• Own hardware lifecycle: procurement specifications, rack-and-stack coordination, BIOS/firmware tuning, and decommissioning.
• Maintain and evolve IPMI/BMC and out-of-band management infrastructure for remote server administration.
• Implement and enforce server access controls, SSH key management, PAM policies, and audit logging in compliance with firm security standards.

Ultra-Low-Latency Linux Tuning

• Tune Linux kernels for deterministic, low-jitter performance: CPU isolation (isolcpus, irqbalance), NUMA-aware process pinning, and tickless (nohz_full) operation.
• Configure and optimize kernel bypass networking stacks (DPDK, Solarflare OpenOnload, Mellanox VMA/DOCA) for sub-microsecond wire-to-application latency.
• Profile and eliminate latency outliers using perf, ftrace, cyclictest, eBPF tooling, and custom instrumentation.
• Build and maintain custom kernel builds with latency-focused patches (PREEMPT_RT, transparent hugepage tuning, CFS scheduler modifications).
• Optimize memory subsystems: hugepage allocation, NUMA balancing, TLB shootdown mitigation, and cache-line-aware memory mapping.
• Tune NIC drivers, ring buffer sizes, interrupt coalescing, and RSS/RFS configurations for maximum throughput with minimal latency.

DevOps & Automation

• Design and maintain CI/CD pipelines for infrastructure-as-code deployments, system image builds, and configuration rollouts.
• Develop automation tooling in Python and Bash for server health monitoring, capacity planning, and incident response runbooks.
• Manage containerized workloads (Docker, Podman) for non-latency-critical services alongside bare-metal trading systems.
• Maintain centralized logging (ELK/Loki), metrics collection (Prometheus/Grafana), and alerting infrastructure.
• Drive infrastructure-as-code practices using Terraform, Packer, and GitOps workflows for reproducible environments.

Network & Connectivity

• Collaborate with network engineers on switch configuration, VLAN design, multicast tuning, and PTP/PPS time synchronization.
• Manage market data feed infrastructure including kernel-bypass capture, gap detection, and feed handler deployment.
• Troubleshoot end-to-end latency across the full stack from exchange NIC to application using hardware timestamping and packet capture analysis.

• 5–10 years of hands-on Linux systems engineering experience, with at least 3 years in a latency-sensitive environment (HFT, electronic trading, or real-time systems).
• Deep expertise in Linux internals: kernel scheduling, memory management, interrupt handling, and the networking stack (from driver to socket).
• Proven experience with kernel bypass technologies (Solarflare OpenOnload, DPDK, or Mellanox VMA) and hardware timestamping.
• Strong proficiency with performance profiling tools: perf, ftrace, bpftrace, eBPF, flame graphs, and hardware performance counters.
• Solid understanding of x86-64 architecture, CPU microarchitecture (cache hierarchy, branch prediction, NUMA topology), and their impact on software performance.
• Production experience with configuration management (Ansible preferred), infrastructure-as-code, and CI/CD pipelines.
• Expert-level Bash and Python scripting; familiarity with C/C++ for reading kernel source and writing low-level tooling.
• Strong knowledge of networking fundamentals: TCP/IP, multicast, PTP, ARP, and switch architecture.

• Experience building custom Linux kernels with PREEMPT_RT patches or contributing to upstream kernel development.
• Familiarity with FPGA-accelerated networking (Xilinx Alveo, Intel PAC) and SmartNIC offload architectures.
• Background in financial exchange connectivity (FIX, ITCH/OUCH, CME MDP3, OPRA) and market data infrastructure.
• Experience with bare-metal provisioning at scale (Foreman, MAAS, or custom PXE/iPXE pipelines).
• Knowledge of security hardening: SELinux/AppArmor, CIS benchmarks, and compliance frameworks relevant to financial services.
• Contributions to open-source projects in systems, networking, or performance tooling.
• Experience in HFT is a hard requirement.

• Highly competitive base salary plus performance-based bonus tied directly to firm and team results.
• Work with some of the fastest trading systems in the world—your optimizations have measurable, real-time impact on PnL.
• Access to cutting-edge hardware: latest-generation CPUs, NICs, switches, and FPGA platforms before they hit general availability.
• Small, elite engineering team with minimal bureaucracy and direct collaboration with researchers and traders.
• Comprehensive benefits including medical, dental, vision, 401(k) with match, and generous PTO.
• Annual learning budget for conferences (P99 CONF, Linux Plumbers, KubeCon), certifications, and hardware lab experimentation.