Note: The job is a remote job and is open to candidates in USA. The hiring client is unnamed and is building a cloud platform for high-throughput, compute-heavy AI workloads. The Staff SRE will own production reliability end-to-end, improve Linux and network performance in a bare-metal environment, and build automation that reduces toil and incidents.
Responsibilities
- Define SLIs/SLOs, run error budget conversations, and ship changes that reduce incidents and improve latency (p95/p99)
- You will build automation to kill toil, improve deployment safety (canary/rollback), and turn observability into signal rather than noise
- You will work close to the metal on Linux performance (CPU/memory/IO contention), kernel tuning, and network debugging (DNS/TCP/TLS, packet loss, congestion)
Skills
- Extensive, recent hands-on experience running Linux on bare metal in data centres, ideally across hundreds or thousands of physical servers, not public cloud only
- Deep Linux expertise, including kernel tuning and troubleshooting performance and stability issues at OS level
- Strong coding skills in Python or Go, beyond automation scripting, and comfortable in a live coding interview
- A track record of building infrastructure automation and tooling that runs at scale
- Ownership of production incidents, from root cause analysis to lasting engineering fixes
- Strong understanding of networking (DNS/TCP/TLS, latency, packet loss, congestion) and experience in low latency environments such as trading, HPC, gaming or large-scale tech
- Kubernetes on bare metal
- InfiniBand
- C development
- Configuration tooling such as Terraform, Puppet or Ansible
Benefits
- 401(k)
- Disability insurance
- Paid maternity leave
- Medical insurance
- 100% remote (United States)
Company Overview