Lead Site Reliability Engineer

Job description

At Zego, we're on a mission to do the good thing, not the insurance thing.

Insurance hasn't changed much in over a century. The way we live, work and travel has. We're building the real-time, AI-driven infrastructure that powers innovative insurance, so good drivers get cover that works the way they actually drive.

For us, AI isn't a line on a roadmap or a buzzword on a slide. It's our operating reality, and it's how we build products that back drivers instead of the old insurance playbook.

Responsibilities

• You will be our dedicated SRE, working alongside Systems Engineering and embedded with Product and Engineering across roughly 120 engineers. Teams own their services and their own on-call. You own the framework, the instrumentation and the AI tooling that makes that ownership work.
• You will make teams good at running what they build: alert quality over alert volume, runbooks an agent can execute rather than prose that describes, and diagnostics good enough that an engineer or an agent reaches resolution without an SRE in the room.
• You will be the primary advocate for reliability with Product and Engineering, making the case with data rather than assertion, so platform health is prioritised alongside delivery commitments.
• Define and land the reliability standard: SLIs, SLOs, error budgets and production readiness criteria that teams apply to their own services, with adoption measured rather than assumed.
• Encode standards into automation rather than enforcing them by hand. Guardrails in CI, agents that check reliability and observability posture on pull requests, and safe defaults in shared infrastructure so the reliable path is the easy path.
• Build SRE capability on Zego's AI platform, extending our MCP servers and agents, and making the estate legible to them through machine readable runbooks, structured telemetry and diagnostics an agent can act on.
• Own observability as a practice, including instrumentation design, signal quality, cardinality and cost.
• Raise the standard of incident response end-to-end, from detection and triage through to retrospectives that produce change. Put AI to work where it pays off most, stripping toil out of incidents so responders can focus on judgement
• Measure toil, publish it, and remove it through tooling that others can run without you.

Requirements

• Deep SRE experience: SLI and SLO definition, error budget management, and ownership of incident response through to blameless retrospectives.
• Fluency with AI as an engineering tool rather than a chat window: building agents and automation, working with MCP servers or equivalent, and judgement about where an agent can act and where a human stays in the loop.
• Strong software engineering in Python, writing tested, maintainable code that other engineers can run and extend.
• Observability depth with OpenTelemetry and Datadog.
• AWS at scale, and Kubernetes based platforms managed through Infrastructure as Code.
• Istio, Crossplane and ArgoCD. We run all three and will teach the specifics to the right person.
• Running ML or AI workloads in production, such as model serving or feature pipelines. We do this at Zego and you would help support it.
• Having been a sole delivery owner somewhere, and knowing what that does and does not mean.