This hands-on workshop bridges the gap between data science experimentation and production ML systems.
Using a real-world problem participants will build out a complete MLOps pipeline step by step. Starting from an exploratory notebook, we’ll add experiment tracking with MLflow, compare and tune models (Random Forest, XGBoost), register and version models in a Model Registry, run batch inference, and set up monitoring for feature and inference drift.
By the end, you’ll understand the full lifecycle and the key questions you need to answer at each stage before going to production.
📋 Pre-requisites for attendees:
👥 Who should attend:
Platform engineers, data engineers, and data scientists who are new to MLOps or want a structured understanding of the full ML production lifecycle
🧠 What they should know:
Basic Python proficiency (comfortable reading/writing Python scripts)
Familiarity with basic ML concepts (what a model is, training vs. inference, metrics like $RMSE$)
A laptop with Git, Python 3.12+, and uv installed (setup instructions provided)
✨ Nice to have but not required:
Experience with pandas, scikit-learn, or any ML framework
🗓️ Agenda:
9:30 AM – 9:45 AM – Registration & Networking
9:45 AM – Opening Remarks & Event Kickoff
11:00 AM – 11:15 AM – Coffee Break & Networking
12:30 PM onwards – Lunch & Networking
