Otto — branded as OttoML — is an open-source chat application that turns machine learning into a natural-language experience for beginners. Published on GitHub under the MIT license with about 978 stars, Otto won third place at the 2020 Facebook AI Challenge hackathon series for its innovative use of the Wit.ai NLP platform. The project was last updated in 2023 and stands as an early example of conversational, no-code machine learning.
Otto assumes minimal familiarity with machine learning. Users simply describe their end goal in plain language and receive intelligent recommendations, or they can pick a sample dataset and start experimenting instantly. Behind the friendly chat interface sit real ML tools: models for regression, classification, and natural language processing, plus preprocessors tailored to the problem at hand. Otto guides users through model selection, shows insightful visualizations of their data and results, and can generate ready-made Python code so beginners go from idea to implementation while actually learning what is happening under the hood.
Otto’s interactions are charmingly direct: tell it you want to label flower species by petal length and it prefills a pipeline and visualizes a nearest-neighbors classification on the Iris dataset; ask it to detect fraudulent credit card activity and it recommends models and opens an interactive neural network designer. These guided, visual moments are the point — Otto teaches the shape of a machine learning workflow by letting beginners drive one with words instead of code.
Otto is completely free and open source under the MIT license.
Students, aspiring ML engineers, and curious beginners who want to learn machine learning by doing — describing goals in plain English and getting working models, visuals, and code back.