Alex Finn demonstrates a terminal interface that groups projects into spaces and lets him switch among coding agents working on different machinesAn AI coding agent is a tool-using AI system that can inspect, modify, and validate software within a repository.. He describes assigning separate agentsA multi-agent system contains multiple AI agents that interact, coordinate, divide work, or influence one another while pursuing tasks. to gameplay, maps, models and testing.
Alex Finn builds a small game with a primary agent, tests the initial result, then assigns parallel changes to other agents. He reviews their progress and tests the combined resultEvaluation is the systematic assessment of a system's behaviour or results against explicit criteria..
Alex Finn also describes custom interfaces for checking and messaging agents from a desktop overlay and a phone. His claims about efficiency reflect his personal experience.
Watch on YouTube
