Build a chess game against a language model
An early tester asked Devin to build a chess game against a language model; it made good progress but eventually hung.
Done withDevin
- What they did
- Early tester Andrew Kean Gao, a computer science student, gave Devin a chess game to build in which the player competes against a language model. Devin works in an isolated environment with a shell, code editor and browser. The source gives no further detail on the steps it took or on how Gao intervened.
- How it went
- Devin made remarkable progress, but at some point it hung and the game wasn't finished. Gao's other tests had mixed results: a Chrome extension worked, and an Antarctic temperature visualization was unsatisfactory.
- Worth knowing
- Devin was only available to selected developers through a waiting list at the time, and the source gives no cost or time figures for this task.
Try it yourself with Devin
Build a chess game where I play against a language model through [API or model name]. It should show the board, enforce legal moves, and call the model for its replies. You're done when I can play a full game from start to checkmate. Tell me if you get stuck instead of looping, and describe what's left unfinished.
Source: the-decoder.com · Mar 13, 2024
Five of these in your inbox every morning
The best things people got an AI agent to do, each with the prompt to try it.
More like this
Hand narrow code changes to Devin for pull requests
A team used Devin for a month on small fixes like chart formatting and API edge cases, with tests and PRs; a complex full-stack task went poorly.
Generate internal dashboards and data migration scripts
A backend engineer says their team uses Devin for internal dashboards and migration scripts, treating output as a starting point that saves about four hours of boilerplate.
Click through apps to verify its own work
With computer use, Devin built and played a desktop game, ordered on Amazon, and tested its own work by clicking through the app.
Play RollerCoaster Tycoon
Ramp Labs put Claude Code in RollerCoaster Tycoon.
Play Doom with computer-use agents as an informal benchmark
The OpenClaw team used computer-use agents to play Doom on remote machines, treating game performance as an informal way to improve the computer-use workflow.
Play a button-clicking game via a browser agent
A developer wired Gemini's Computer Use model to Playwright in Java and had it play a simple button-clicking game to verify click-coordinate scaling.