Your first machine learning model. In 40 minutes.

Mariana Antaya is a former AI Product Manager at Microsoft turned full-time founder and creator. She built AI agents and evaluation frameworks for enterprise products there, and now ships her own maching learning (ML) models and teaches a community of 700K+ how to apply AI to their work, hobbies, and interests.

In this session, Mariana breaks down her full 5-step ML workflow — taking raw, messy e-commerce data and turning it into a customer prediction model that actually answers a business question: will this customer buy again in the next 90 days? Most ML tutorials hand you clean Kaggle data and a model choice, then call it a day. The model itself is maybe 10% of the work. The other 90% is cleaning, feature engineering, defining the target, and reading the output. That is the part nobody teaches, and it's the part that separates a tutorial from a portfolio project.

This session is for creators, professionals, and aspiring ML builders who've watched 50 AI videos and still can't build anything real. You'll walk away with a working model, the code behind it, and a workflow you can reuse on any dataset.

What you'll learn:

• The 5-step ML workflow: clean → engineer → target → train → interpret
• How to clean real-world, messy data without losing the signal that matters
• The RFM framework (Recency, Frequency, Monetary) - three numbers that capture almost everything about a customer
• How to define a target your model can actually learn - the silent killer of most beginner projects
• A trained Random Forest model and the code behind it, ready to drop into your portfolio