Wat is GraphQL?

GraphQL was developed by Facebook (Meta) and released publicly in 2015 as an alternative to REST APIs. While REST uses multiple endpoints with fixed data structures, GraphQL uses a single endpoint where the client specifies exactly what data it wants using a query language.

The core advantage is flexibility. With REST, if you need a user's name, email, and their last 5 orders, you might need to call /users/123 and then /users/123/orders — two requests that return more data than you need. With GraphQL, you send a single query that returns exactly those fields and nothing more.

GraphQL uses a typed schema to define all available data and operations. This schema serves as a contract between frontend and backend teams, automatically generates documentation, and enables powerful development tools with autocomplete and validation.

The tradeoffs include more complex server implementation, potential performance issues with deeply nested queries, challenges with caching (since every request can be unique), and a steeper learning curve. GraphQL shines when you have multiple client types (web, mobile, IoT) that need different subsets of the same data, or when over-fetching and under-fetching from REST endpoints becomes a problem.