Infrastructure Engineer / SRE

The role

You will own the physical and logical infrastructure that keeps our trading systems running with zero unplanned downtime during market hours. This includes co-location server management, network topology and cross-connect provisioning, Linux kernel tuning for latency, and the observability stack that enables the engineering team to diagnose issues at nanosecond resolution.

This is a hands-on role — you will rack and cable servers, configure switches, write Ansible playbooks, and be on-call when something breaks. There is no separation between "infra" and "ops" here.

Responsibilities

• Manage server hardware at co-location facilities: provisioning, hardware fault isolation, component replacement coordination.
• Own network topology and cross-connect infrastructure; maintain routing tables, VLAN config, and BGP peering with exchange feeds.
• Tune Linux systems for low-latency operation: CPU isolation, IRQ affinity, NUMA binding, huge pages, transparent huge page disabling.
• Maintain PTP clock synchronisation infrastructure across all co-located and office systems.
• Build and maintain the observability platform: metrics collection, alerting, latency dashboards, and post-trade analysis pipelines.
• Automate deployment and configuration management using Ansible and custom tooling; maintain infrastructure-as-code discipline.
• Participate in incident response and post-mortem process; write detailed root-cause analyses.

Requirements

• 5+ years of Linux systems administration or SRE experience in a performance-critical environment.
• Deep knowledge of Linux networking: tc/qdisc, ethtool, kernel networking stack, NIC driver tuning.
• Experience with data-centre networking: switch configuration, VLAN design, LACP, BGP basics.
• Proficiency with infrastructure automation: Ansible, Terraform, or equivalent.
• Solid scripting ability in Python and Bash; comfort reading C or Go is a plus.
• Experience with time-series observability stacks (Prometheus, InfluxDB, Grafana, or custom solutions).
• Ability to read kernel traces and system-level profiling output (perf, ftrace, eBPF).

Nice to have

• Prior experience at a co-location facility, exchange, or high-frequency trading firm.
• Familiarity with DPDK or RDMA network stacks from an operational perspective.
• Experience managing PTP/IEEE-1588 clock synchronisation in production.
• Knowledge of FPGA PCIe device management and driver operations.