AI Platform Engineer

Centrical runs a production platform serving global enterprise customers at scale. Inside it, we are building a new operational model: instead of a specialist for every domain, we build AI systems that do the operational work: monitoring, diagnosing, and increasingly fixing production issues on their own. We have already automated one full operational function this way, end to end, running in production today. This role exists to take that model across the company.

The center of gravity of this role is building, not firefighting. Production issues will reach you, and each one is raw material: you investigate it once, deeply, then build the agent, the automation, or the gate that handles the next one without you. The role gets quieter as your systems get smarter, by design.

This is an AI-first engineering role. Building with LLMs and agents is not an occasional tool here; it is the primary way the work gets done, every day.

What You’ll Do

• Build the automation layer for our production estate: AI agents that analyze incidents and propose fixes, dashboards that surface problems before customers feel them, alerts that mean something, and automations that execute routine operational work safely.
• Build guardrails into the delivery pipeline so performance regressions are caught before they reach production, not after
• Extend our existing AI operations agent with new capabilities and roll it out to more teams, with the approval flows and safety design that production automation demands.
• Investigate interesting production problems end to end, from a database execution plan through service logs down to the application code, and turn what you learn into a permanent capability.
• Work closely with teams across the company: R&D, Data, DevOps, and Customer Success. Every automation you build starts from understanding how another team works, and every domain you enter comes with the people who know it best.
• Expand into whatever domain the next problem lives in: relational and non-relational databases, event pipelines, caching, cloud infrastructure. The scope of this role grows with you.
  
  

• 3+ years as a backend, infrastructure, or DevOps/SRE engineer, with real production experience.
• Proven hands-on experience building with LLMs and AI agents: real automations or internal tools in use, not experiments. This is the core requirement of the role and the first thing we will ask about.
• Solid SQL and genuine willingness to go deep into database internals.
• Ability to read, understand, and fix code in a large existing codebase.
• An automation reflex: recurring manual work should bother you until it becomes a script, a pipeline, or an agent.
• Hunger to learn. This role means working with many teams and many technologies: you will touch more of a production stack, and more parts of the company, in your first year here than most engineers see in five.
  
  

• Relational database tuning at scale (execution plans, indexing, concurrency)
• Non-relational data stores and message-driven architectures
• Observability platforms and alert design
• Cloud infrastructure and CI/CD ownership
• Multi-tenant SaaS environments