SRE & Observability Engineer
SLOs that mean something, alerting that stays quiet unless it counts, and on-call that stops being a mystery.
Full-time / engagement Hybrid - Modi'in & client sites across Israel 5+ years, SRE focus
What you'll do
- Define SLOs and error budgets tied to user impact - not vanity dashboards
- Instrument services (OpenTelemetry), build Prometheus/Grafana/Loki stacks and kill alert noise
- Run incident reviews and write postmortems that change behavior
- Design on-call rotations, runbooks and escalation for clients
- Build cost-aware monitoring: metrics-vs-logs decisions that keep the bill sane
- Apply the same discipline to AI adoption client work (agent workflows, safe access, measurable outcomes)
Who we're looking for
- Real production SRE experience - incidents you've seen through, not just dashboards you've built
- Fluency in the classic stack: Prometheus, Grafana, Loki/ELK, OpenTelemetry
- Coding ability in at least one of Go/Python for tools and instrumentation
- A calm, structured approach to fires - you document the fix while fixing it
- Opinions about when NOT to add another alert
Why this works here
- SRE done properly at companies where uptime is the product
- You'll shape our AI observability practice (LLM traces, evals, cost) as it grows
- Teaching on-call discipline and runbook culture to client teams
- A clear path: own the observability pillar across every engagement
Sound like you?
Send a short note about the systems you've run and what you've built - we read everything, and a senior engineer replies (not HR).