Top 10 Rest API Design Pitfalls
summary
Collected from work with over 150 companies, this talk lists concrete API mistakes and fixes: avoid breaking changes by keeping optional inputs, use UUIDs instead of sequential IDs, cap batch requests (avoid 100,000-item pulls), and prefer server-side aggregation or async CDC over synchronous data dumps. Practical tools and patterns: compare OpenAPI extracted from code with the Git-stored spec (OpenAPI diff (extract from code vs git)), find callers via trace ids / OpenTelemetry, enforce contract sync with Spring Cloud Contract or client libs plus dependabot/renovate PRs.
AI-written summary · about
chapters (8)
- 0:00Introduction and speaker background
- 4:45Backwards compatibility pitfalls
- 15:13Performance and batch operations
- 23:20Scalability and maintainability mistakes
- 31:04CQRS and data consistency
- 42:43URL design and HTTP methods
- 47:21Error handling and API responses
- 50:57Summary and key takeaways
related talks
52mTop 10 Rest API Design Pitfalls
49mTop REST API Design Pitfalls
52mTop REST API Design Pitfalls
58m15 Best Practices for Designing Enterprise RESTful APIs in the Cloud
27mAddressing OWASP API Security Top 10 starts at design time
42mREST API design good practices
18mBuild a terrible API for people you hate
01h 06mGood API design