Define and champion non-functional requirements (NFRs) and availability targets for services/products in electronic colo trading.
Ensure reliability, performance, and operational excellence of Linux-based compute platforms; emphasize automation, standardization, and disciplined incident management.
Collaborate with network, trading technology, and data-center teams to optimize latency, throughput, and stability (kernel/IRQ/CPU isolation, NUMA, NIC tuning).
Develop and track service level indicators/SLIs and service level objectives with stakeholders; establish and manage error budgets.
Operate and improve observability across the stack (metrics, logs, alerting); translate signals into actionable runbooks and improvements.
Lead major incidents (triage, mitigation, root-cause analysis, corrective actions) and coordinate hardware-adjacent colo tasks.
Leverage AI capabilities to accelerate triage and post-incident analysis; validate outputs and ensure security/compliance; mentor engineers.
Technical Stack
必須スキル
Bachelor’s degree and 5+ years in Site Reliability Engineering or related field; formal SRE training/certification
Production Linux administration (RHEL/Debian/Ubuntu); strong OS, hardware, and basic networking troubleshooting
Linux internals, performance tuning, and diagnosing complex latency/reliability issues
Automation with Bash and Python or Go; configuration management, CI/CD, monitoring, observability
Incident management, root-cause analysis, and postmortems; drive durable improvements
Use of enterprise AI capabilities with validation, guardrails, and data sensitivity awareness
歓迎スキル（該当する場合）
Experience with electronic trading, latency-sensitive environments, or colocation data centers
Kernel/CPU pinning, IRQ tuning, time synchronization, and low-latency Linux optimization
Infrastructure-as-code patterns and standardized “golden” server builds; fleet management and patching
Experience operating at scale with automation-driven patching and drift control
Familiarity with AI-assisted reliability workflows across SDLC (CI/CD checks, runbooks, validation)