Designing Resilient APIs
Contracts, versioning and graceful failure for services people depend on
By Omar El Alaoui
One-time purchase
€32.00
VAT included
where applicable
- Download starts the moment you pay
- Reads anywhere a PDF does
- Edition updates come free
- Checkout handled securely
About the book
What you’re getting into
An API is a promise, and every client you win makes that promise harder to change. This book is about keeping it: request and response contracts that survive real-world mess, versioning that never strands the people downstream, and failure modes clients can actually recover from. By the end you can design an interface teams trust enough to build their own work on top of.
After you’ve read it
What you’ll be able to do
- Shape resource and RPC-style contracts that age gracefully
- Choose status codes and error bodies a client can act on
- Evolve an API without breaking existing callers
- Make endpoints idempotent so retries are safe
- Apply timeouts, retries and backoff without amplifying an outage
- Add pagination, filtering and rate limits that hold up at scale
- Write docs and contract tests integrators actually rely on
Table of contents
- 01
An API Is a Promise
- The true cost of a breaking change
- Consumers you'll never meet
- A design workflow for the book
- 02
Modelling Resources and Actions
- Resource-oriented vs RPC thinking
- Naming, nesting and granularity
- When strict REST stops helping
- 03
Errors Clients Can Recover From
- Status codes with intent
- Structured, machine-readable error bodies
- Retryable versus terminal failures
- 04
Versioning Without Stranding Anyone
- URI, header and media-type versioning
- Additive change and deprecation windows
- Sunsetting an old version safely
- 05
Idempotency and Safe Retries
- Idempotency keys in practice
- Exactly-once as a client contract
- Deduplicating writes on the server
- 06
Timeouts, Retries and Backpressure
- Budgeting a request's time
- Backoff, jitter and retry storms
- Shedding load on purpose
- 07
Pagination, Filtering and Limits
- Cursor versus offset paging
- Query parameters that scale
- Rate limits, quotas and fairness
- 08
Documentation Integrators Trust
- The spec as source of truth
- Examples over prose
- Contract testing across teams
The full chapter list, exactly as it appears in your PDF.
More in Backend Engineering
