June 2023 Summaries

Candid, a medical billing API provider, has adopted Fern's schema-first API development approach. This method reduces the time developers spend on repetitive tasks such as writing boilerplate code and managing multiple sources of truth for their API. By using Fern, Candid generates Pydantic models, FastAPI routes, TypeScript clients, and auto-generated lines of backend and frontend codes. The company plans to use Fern for syncing API definitions with API docs and generating SDKs for customers in the future. This adoption has significantly improved efficiency and minimized errors caused by out-of-sync API components.

Candid, a company offering an API for medical billing, faced challenges with repetitive code writing and synchronization issues across multiple sources of truth, leading to runtime bugs and inefficiencies. By adopting Fern's schema-first API development approach, Candid streamlined its processes significantly. Engineers first define schemas using yaml files similar to OpenAPI, then generate necessary backend components like Pydantic models and FastAPI routes with Fern, which integrates with mypy for compile-time error checking. Once business logic is implemented and merged, Fern automatically generates a TypeScript client for the frontend, ensuring consistency and reducing boilerplate code. This approach has resulted in 95,000 lines of backend and 30,000 lines of frontend code being auto-generated, minimizing wasted engineering time and establishing a single source of truth for the API. Candid plans to further enhance its operations by using Fern to keep API documentation in sync and to generate customer-facing SDKs, improving integration efficiency.