Senior Software Engineer – GraphQL Federation Platform (Apollo Router & Backend Services)

Job Title : Senior Software Engineer – GraphQL Federation Platform (Apollo Router & Backend Services)

Location : Sunnyvale, CA

 

 JD :

Hybrid role

Job Title: Senior Software Engineer — Federated GraphQL Platform (Router & Services)


Role Description:

About the role We run a federated GraphQL data layer — a middleware tier that composes many back-end services into a single, typed API that powers large-scale, consumer-facing web experiences.
At the center of that layer is a federation router that plans and executes queries across dozens of subgraphs; around it sits the services (subgraphs) that own and expose each domain of data.
We’re looking for a senior engineer to own the hardest, highest-leverage parts of that platform: the router strategy (adopt, extend, or build) and the core services our team operates, while helping the broader org build on the graph.
You’ll work close to the metal on performance and correctness, and close to the roadmap on where the platform goes next.
What you’ll do Own the router strategy — adopt, extend, or build. Evaluate whether the open-source Apollo Router meets our needs for performance, extensibility, and operability, and build the delta features we require on top of it (custom plugins/coprocessors, auth, caching, observability, query governance).
Where the open-source path falls short, lead the decision to either build a purpose-built router in Rust or adopt and extend an alternative open-source router (e.g. Hive Router / Hive Gateway, Cosmo, or similar).
Prototype, benchmark, and make the call with data. Define the router’s non-functional bar: latency, throughput, resilience, safe schema composition, and zero-downtime rollout.
Own and operate core services (subgraphs). Own the subgraphs our team is responsible for — including the Site Config service and related domain services — end to end: schema design, resolvers, performance, reliability, and on-call.
Keep these services exemplary: clean federated schemas, strong contracts, good tests, and sensible SLOs.
Enable other service teams. Support partner teams building their own subgraphs — across stacks (Node/TypeScript, Scala, Java/Spring) — review federated schema designs, share patterns and tooling, debug cross-service query issues, and raise the overall quality and consistency of the graph.
Reduce the cost of joining the graph: paved paths, templates, docs, and guardrails. Advance the platform roadmap.
Help shape and deliver the multi-quarter roadmap for the data layer — turning architectural direction into shipped, measurable improvements. Bring a point of view on where federation tooling is heading and how we should invest.