{"type":"project","slug":"fieldwork-ml-studio","title":"Fieldwork (ML Studio)","status":"active","tagline":"Learn machine learning by preparing data, training models and testing predictions.","summary":"A guided browser studio for tabular machine learning, with four learning labs, model comparison, an experiment notebook and optional device saves.","canonicalUrl":"https://jonathantipper.com/projects/fieldwork-ml-studio","markdownUrl":"https://jonathantipper.com/projects/fieldwork-ml-studio.md","jsonUrl":"https://jonathantipper.com/projects/fieldwork-ml-studio.json","launchUrl":null,"launchLabel":null,"launchNote":null,"started":null,"tags":["machine-learning","learning","browser"],"links":[],"image":{"src":"https://jonathantipper.com/images/projects/fieldwork-ml-studio-preview.webp","alt":"Fieldwork data exploration workspace using synthetic customer records","mode":"screenshot"},"content":"## What it is\n\nFieldwork is a browser studio for learning machine learning on tabular data. Start with a guided example or import a table, then prepare the data, train a model and inspect its predictions.\n\n## What it does\n\nFour labs cover customer retention, revenue estimates, shopping patterns and unusual equipment readings. An experiment notebook keeps validation scores, baselines and reflections together. Training and predictions run in the browser; saving a project on the device is optional, with downloaded backups available.\n\n## Current status\n\nAn active learning studio. The app's learning workflow needs no model API key. It is designed for bounded experiments on tabular data, with documented size and evaluation limits."}