You will build products that automate and productize platform operations. Your first product is the unified service catalog: the system that records what services we run, who owns them, how they are deployed, and how they relate to each other, and that security and reliability partners and development teams use as their working view of the platform. You will build and extend the catalog, integrate it with our source control, delivery, infrastructure, and observability systems, and expose it through APIs, a CLI, and chat and web interfaces, including AI-assisted ones.
The catalog is the foundation for a longer roadmap (self-service deployment, workflow automation, and AI-assisted operations) that you will contribute to as it develops. The catalog is security-sensitive, so implementing authorization, participating in threat modeling with security partners, and producing compliance-ready audit evidence are part of the job. You will use AI coding agents as a core part of your daily workflow and apply good judgment about when an LLM is the right tool and when deterministic code is.
● Build and extend the unified service catalog: implement the data model, APIs, and sync jobs that bring ownership, deployment, dependency, and compliance state from our source control, delivery, infrastructure, and observability systems into one queryable system.
● Build integrations: webhooks and event-driven processing, idempotent reconciliation against external systems, and clean API contracts that other teams can build on.
● Expose the catalog through a CLI, chat, and web interfaces, and contribute to the AI-assisted interfaces on top of it, such as assistants that answer questions about services and ownership. Agents are designed to run under least-privilege identities, with auditable actions, and prompt injection is a first-class design consideration.
● Implement the authorization model that security partners specify, controlling who can see and change catalog data across teams.
● Produce compliance-ready audit evidence (ownership, change, and access records) as a built-in output of the catalog.
● Participate in threat modeling with security partners and implement the resulting controls: least-privilege access, secrets hygiene, and policy checks in the catalog.
● Ship production-quality software: tests, CI, observability, and runbooks for what you build.
● Instrument what you ship: adoption, data freshness, and manual request volume for the catalog.
● Write clear documentation and support the teams that consume the catalog.
Participate in a light on-call rotation for the catalog. This is not a production streaming on-call.