Inside the Conversation
In this live Customer Spotlight interview, Dan Boerner sits down with Jonathan Wondrusch to look at how Block is preparing for the next wave of agent driven development. Years ago Block bet on GraphQL and Apollo, and that choice paid off quickly by giving teams a scalable alternative to monolithic APIs.
Now that same foundation is paying off again as Block shifts to MCP. The graph gives AI agents a clean and predictable interface, but the scale is intense. Thousands of API endpoints need to be brought into the graph, and doing the work manually would take far too long.
Join to see how they used their AI assistant Goose and Apollo MCP Server to auto-generate Apollo Connectors definitions in a fully automated workflow. The result is a fast, repeatable path to MCP access that brings their entire API surface into the graph without extra work from engineering teams.
Interview Topics
Declarative automation: How AI generated connectors bring APIs into the graph at scale.
Agent interface design: Why the federated graph gives MCP agents a stable and predictable contract.
Operational safety: Adding guardrails to agent driven workflows.
Platform advantage: How Block’s graph foundation gives them a faster path into AI.
Meet your speakers
