AI writes your code in seconds. Then you wait twenty minutes for CI to tell you whether it works. The bottleneck has moved: writing code is cheap, but verifying it against a slow build is now what stands between you and fast feedback.

The biggest lever for faster Spring Boot builds is a hidden gem most teams barely use: Spring Test's context caching. Done right, your integration tests reuse expensive application contexts instead of starting fresh ones. Done wrong, every test class silently pays for a new context — and nobody notices until the pipeline takes an hour.

This talk introduces the Spring Test Profiler, an open-source tool that makes context caching visible. We'll start with the theory: how the cache works, what goes into the cache key, and why seemingly harmless choices — @MockitoBean, @DirtiesContext, dynamic properties — quietly multiply your build time. Then we get practical: using the profiler to spot cache misses and applying proven fixes, drawn from years of optimizing test suites across many teams, that cut build times by up to 70%.

We'll also cover the pitfalls: "optimizations" that break test isolation or hide flakiness. A fast build that lies to you is worse than a slow one.

This session assumes you already write tests and know your way around a CI pipeline. You'll leave with a concrete, prioritized checklist to reclaim hours of feedback time in your own project.