Article: The case for boring infrastructure

New infrastructure is exciting precisely because nobody has been burned by it yet. Boring tools earned their reputation by failing in well-understood, well-documented ways that someone has already solved.

Every novel component you add is a new on-call surface and a new thing to explain to the next engineer. The cost is rarely visible on the day you adopt it and very visible the night it breaks.

Save your innovation budget for the parts of the product that are actually your edge. Let the database, the queue, and the proxy be aggressively, gloriously dull.