Has AI made it too easy to keep building instead of validating?
disclaimer: I’m guilty of this.
The more agentic capability i seem to have at my disposal, the deeper i go. But is that where I should be?
Agents are running, code keeps appearing, and every prompt uncovers three more things that could be improved.
Another refactor. A cleaner interface. One more feature. A few more edge cases.
Because building is now so fast, it feels almost wasteful to stop.
When I started building startups around ten years ago, I relied heavily on the Design Sprint approach created by Jake Knapp at Google:
Form a hypothesis around a real problem.
Find the quickest way to test it.
Execute, measure and learn.
Take the evidence into the next sprint.
Each sprint was something that started Monday - closed Friday to analyse and form next iteration.
Today, I can easily spend a whole day perfecting something nobody has asked for.
The product may be getting 'better', but I’m not necessarily getting any closer to knowing whether somebody wants it.
AI has changed how quickly we can build. It hasn’t removed the need to put something imperfect in front of a real person and see what happens.
Hypothesis. Smallest test. Evidence. Next sprint.
Anyone else guilty of staying in the build for too long because the agents make it so easy?
Replies