A complete, progressive path from the basic vocabulary to a model in production. Problem framing, data preparation, classification and regression algorithms, metrics and business costs, leak-proof pipelines, interpretability, deployment and monitoring — with hands-on labs throughout.
Built to make you work, not listen: five-minute lessons, a quiz after each one, guided exercises on one dataset of 40 industrial machines. From pandas, SQL and feature engineering to regression, random forests, clustering and neural networks; then embeddings, RAG, local LLMs with Ollama and LangChain, text-to-SQL tools, LangGraph agents and a Streamlit diagnostic app. Free, offline, no API key.
