How to Build End-to-End ML Systems with Genie Code

In April 2026 Databricks announced that for the first time more code was written by Genie Code than by humans on the platform. This session is a deep dive on Genie Code, and specifically how it democratizes machine learning authoring and serving for Data Scientists and ML Engineers, and soon all data practitioners workflows. Genie Code for ML workflows acts as an expert ML copilot that guides even junior data scientists through complex end-to-end ML journeys. Purpose-built for data + AI tasks and with an integrated UX, it enables seamless data discovery, exploratory data analysis, feature engineering, model selection, training and evaluation, hyper parameter tuning and model deployment and monitoring, all while fully aware of your context, metadata and signals. And all this with the governance and lineage of the Unity Catalog. This is where agentic ML is headed when authoring, serving, and operations collapse into a single agentic loop.