From hackathon to beta.
A deadline made Webwoven work. Real players revealed what it would take to turn it into a product.
OpenAI Build Week gave me something very useful for Webwoven: a hard deadline. An idea had to become playable quickly. That forced me to expose the core: a starting point, a target, understandable connections and as few distractions as possible.
AI-assisted development helped me test alternatives and establish foundations faster. The hardest decision still remained human: What belongs in the first build, and what should deliberately wait?
A deadline creates a build
Webwoven did not need to do everything for the submission. It needed to prove its idea. Players move through a knowledge graph and connect real entities through named relationships. Wikidata provides the verifiable foundation. The game turns it into a path people can understand and compare.
A hackathon rewards speed. That is useful as long as a finished build is not confused with a finished product. After the submission, the less visible work began: securing state, sharpening terminology, making errors understandable and shaping the start so that nobody needs an explanation from me.
Feedback creates a beta
The beta added solo rounds, a daily challenge, multiplayer and different knowledge categories. More important than the longer feature list were the first real questions: Which connection was surprising? Where was the interaction unclear? When did a route feel clever, and when did it feel random?
Some feedback leads to a new feature. Other feedback reveals small problems in onboarding, performance or difficulty. Both matter. I try to treat changes as hypotheses, release them deliberately and observe whether they actually improve the game.
Build Week made Webwoven work. The beta taught me what to work on next.