ap / engineering archive
← All field notes

Why successful demos still fail in production

The happy path demonstrates possibility. Production is the accumulation of unusual inputs, interrupted requests, and ambiguous ownership.

Edited · 1 min read · Engineering / systems

The real system begins where the demo ends: timeouts, rate limits, malformed documents, stale sessions, and retries after partial success.

I now treat failure behavior as part of the feature specification. If the system cannot explain its state or recover safely, the workflow is not finished.

A useful review question is: what will the user see if the request succeeds in one store and fails in another? The answer should name a visible state, the owner of the retry, and the cleanup boundary. Adding another spinner does not answer any of those questions.