Turning a private GraphQL API into a public one comes with unexpected challenges. We’ll share how we approached this transition—starting from an existing internal schema that wasn’t shaped for external consumers—and the steps we took to expose only what was ready. Using Apollo Federation Contracts, we filtered out unstable or sensitive parts of the graph. Along the way, we defined best practices for the public schema, like cursor-based pagination, using oneOf for inputs and results. We’ll also touch on how we serve the schema through Hive Gateway with a supergraph setup, and the security measures we added, like depth limiting and complexity analysis. To keep things evolving safely, we rely on GraphQL Hive to track usage and guide deprecations.

If you’re thinking about exposing a GraphQL API—or just want ideas for keeping one clean and manageable—this talk will share what worked for us, what didn’t, and what we learned.

Join Dan from rconnect.tech as he introduces GQLPT, a npm library that transforms plain text into GraphQL queries. Learn how to set up a TypeScript instance of GQLPT and replace your existing GraphQL calls with simple plain text queries. Dan will also dive into advanced topics, showing you how to take GQLPT to production using RAG (Retrieval-Augmented Generation) and vectorized embeddings. This talk is perfect for developers looking to streamline their GraphQL workflow and leverage AI for more efficient API interactions.

Apollo’s response caching has been rebuilt from the ground up for speed, control, and scale—and it’s easier than ever to use. In this session, you’ll see how to flip on supergraph-level caching with directives and instantly cut latency, all without touching subgraph code.

We’ll get into the good stuff: policy-driven caching, surrogate keys for precise invalidations, and patterns to keep data fresh across a federated graph. You’ll see a live demo of the cache debugger in Apollo Sandbox, complete with expanded metrics and distributed tracing so you actually know what’s going on. We’ll also show off the new invalidation engine—it’s dramatically faster, handles more complex scenarios, and is just plain nicer to work with.