AI Development ChallengeCan you build with LLMs, not just chat? Prove it on the leaderboard.
One hour, a real box, and a broken AI feature to wire up. The best scores get it returning correct, grounded answers. Get a score out of 100, a tier, and your rank.
Free to play. Runs in your browser. Sign in to save and share your rank.
Endpoint returning cited answers in 34 minutes. Clean retrieval.
The challenge
Fix the AI endpoint
An answer endpoint calls a model but returns hallucinated, uncited results. This is the actual console you get. Wire retrieval in correctly and get it answering from the real source.
What we score
Outcome, root cause, restraint
Everyone gets the same broken project. The score is your judgement: did it come back to life, did you find the real cause, and was the fix clean.
Get it working
Outcome is graded first. A change that does not run is not a fix, however elegant it looks.
Find the real cause
We reward finding the actual root cause and fixing that, not patching the symptom and hoping.
Keep it clean
A minimal, correct change scores higher than a pile of guesses. Restraint counts.
The leaderboard
Where will you land?
Every run is judged on the same challenge, so the score is comparable. Climb from Learning to Elite, beat your friends, and share a link that shows exactly where you rank.
Claim your rankHow it works
Three steps, one hour, one rank
Boot the broken stack
A real Linux machine spins up in your browser with the broken project, an editor, and a terminal. Nothing to install.
Fix it yourself
You have an hour. Read the logs, find the real root cause, and make the smallest change that brings it back to life.
Get scored and ranked
When it is live, the judge scores your run out of 100, hands you a tier, and drops you on the leaderboard.
Find out if you can really build with AI
One broken feature, one hour, one shareable rank. No setup, no install. Start in your browser right now.
FAQ
Questions, answered
It is a hands-on challenge on a real Linux machine. You fix a broken LLM feature so it returns correct, grounded answers from the real source. You have up to an hour, and we score the outcome, whether you found the real root cause, and how clean your change was.
No. The whole challenge runs in your browser on a real Linux box with an editor and a terminal already set up. Nothing to download, nothing to configure.
You have up to an hour, but the highest scores finish fast and clean. It is free to play. Sign in to save your result and share it.
A shareable score out of 100, a tier, and your rank against everyone else who has taken it. Send the link to a friend and challenge them to beat it.