November 2025 Summaries

Block's GraphQL platform team successfully scaled graph adoption by creatively using Apollo Connectors, typically designed for REST APIs, to integrate their gRPC services into a unified graph without requiring new infrastructure or tools. By leveraging an existing JSON translation layer and open-source AI agent Goose, they established a repeatable framework enabling over 3,000 services across 700 teams to transition seamlessly. The breakthrough occurred during a four-day hack week when teams efficiently integrated APIs into the graph using Apollo Connectors, prompting the platform team to prioritize Connectors as the default integration method. To enhance reliability and scalability, Goose Recipes were developed, standardizing AI-driven connector generation, while the orchestration system Flock facilitated the parallel processing of multiple endpoints, improving success rates and minimizing manual interventions. This systematic approach not only streamlined API integration but also transformed Block's platform development, demonstrating the potential of structured workflows and multi-agent orchestration in modernizing API management.

At the GraphQL Summit 2025, Wiz unveiled how it integrates Apollo MCP with its Model Context Protocol (MCP) to streamline and secure AI-driven developer workflows, particularly in cybersecurity. Wiz's approach automates and simplifies the integration process by embedding API documentation directly into the IDE, allowing AI agents to generate complete integrations with minimal manual input. This system leverages Wiz's WIN API and Apollo MCP to provide real-time, validated queries and payloads, reducing the need for extensive manual testing and ensuring adherence to best practices. By embedding security validation within the development process, Wiz enhances the reliability and security of AI-generated code, addressing the dual challenge of developers and attackers using AI. This strategy, which is applicable to any platform with API-driven integrations, turns API documentation into an interactive interface, promoting speed, security, and scalability in modern development environments.

Athenahealth's adoption of Apollo GraphQL showcased at the GraphQL Summit 2025 highlights a strategic approach to modernizing healthcare systems by ensuring real-time data updates without disrupting existing operations. The initiative was sparked by a critical need to prevent potentially harmful errors, such as outdated allergy information during surgeries, in their clinical software that supports over 20,000 medical practices. Architect Colin Barringer introduced a "steel thread" approach, starting with a low-risk, high-visibility test case in the ambulatory surgical center workflow, which required instant data synchronization that legacy systems couldn't support. By integrating Apollo GraphQL, Athenahealth successfully implemented real-time updates, significantly increasing operation volume and demonstrating both safety and speed. Their platform, treated as a product, emphasizes consistency and developer experience, using tools and templates to maintain safety and velocity. The company is also exploring the safe integration of AI agents into their system, ensuring trust and safety are prioritized. Athenahealth's experience offers a blueprint for modernizing legacy systems in regulated environments, showcasing the importance of a controlled, scalable approach that emphasizes safety and real business value.

At the GraphQL Summit 2025, leaders from Capital One, Expedia, Intuit, and Wayfair discussed how they leveraged AI to enhance developer productivity by reducing manual review loops and governance bottlenecks, using a federated GraphQL architecture to provide the necessary structure for AI initiatives. They implemented AI solutions like instant first-pass reviews, AI-powered IDE plugins, natural language queries, and intelligent routing to speed up processes without undermining human judgment. Wayfair achieved immediate feedback in schema proposals, Intuit integrated governance checks into the IDE, Capital One enhanced approval flow efficiency, and Expedia facilitated conversational runtime insights, collectively addressing the productivity tax unnoticed by finance teams. Each company tailored its approach to AI trust and integration, emphasizing the importance of human oversight in high-stakes situations, with the overarching strategy of using AI to handle volume and context while humans focus on critical decisions. These efforts collectively demonstrated that AI could significantly accelerate platform delivery when implemented with a focus on trust and quality from the outset.

Mastering cache key resolution in Apollo iOS involves using the @typePolicy and @fieldPolicy directives to efficiently handle the storage and retrieval of cached data, thereby minimizing network requests and avoiding data duplication. These directives allow developers to define how cache keys are constructed, ensuring that objects with the same identity are stored consistently and can be retrieved intelligently based on query parameters. The @typePolicy directive specifies how cache keys are derived from fields within the fetched objects, promoting consistent storage, while the @fieldPolicy directive focuses on deriving cache keys from field arguments to facilitate effective data retrieval. This approach not only enhances data consistency and reduces memory footprint but also significantly improves application responsiveness by cutting down unnecessary network traffic. For complex scenarios that require more than the declarative power of directives, programmatic configurations offer additional flexibility, allowing developers to implement custom logic for cache key resolution. By mastering these techniques, developers can optimize Apollo iOS applications for efficient caching, ensuring faster and more reliable user experiences.