Definition
Plain language
A popular online platform that hosts data-science competitions.
As stated in the literature
A platform for machine-learning competitions and datasets, whose competition format is used as a template for autonomous ML-engineering benchmarks like MLE-bench.
Why it matters: Its competition format has become a ready-made template for testing whether AI agents can do machine-learning engineering on their own.
For example, a company might post a Kaggle contest offering a prize to whoever builds the most accurate model to predict customer churn.
Heard on the show
“… experiments on MLE-Bench Lite — that's a benchmark from OpenAI where the agent enters real historical Kaggle machine-learning competitions and gets scored against the human leaderboard. …”Episode 131 — Why Autonomous Research Agents Forget Their Own Lessons, and Arbor's Fix