Dots builds a browser game with four agents, Jev cuts decision time 3.6x
Given a 12-second clip, Dots used four agents to produce a playable browser game, with Jev deciding whether each check passed, failed or needed review, making the workflow about 3.6x faster and 4.2x cheaper than full review.
i gave Dots a 12-second clip and got a playable browser game back four agents handled the build. Jev helped decide what happened next when a check passed, failed or needed another look the brief was one sentence: “rebuild this mechanic as a browser game with keyboard controls, scoring and instant restart” Dots split the work: 1. break down the movement and game rules 2. build the scene and interface 3. implement controls, collisions and scoring 4. play through it in the browser and return failures Jev checked the current stage and test results. routine fixes went back to the builder; unclear failures went to deeper investigation the finished build had: > keyboard controls > working collisions and score tracking > restart without reloading > 24/24 gameplay checks passed I compared two versions of the workflow against the same brief: → Dots reviews every next step: 6h 24m / $82.40 estimated model cost → Jev handles routine decisions first: 1h 46m / $19.60 estimated model cost about 3.6x faster and 4.2x cheaper both builds passed the same gameplay checks. Jev's calls are included in the estimate one reference clip → four coordinated agents → a game I could open and play full guide and setup below ↓
This post quotes another post: x.com/mika_systems/status/2107101689878630885