Winter Term AI Builders Bootcamp: 10-Week Advanced Track

This ten-week winter bootcamp is for teens who already code in Python and want to go beyond using AI tools to actually building them. Across the term, learners work through the full machine learning workflow and finish by shipping a team project they can show to schools, clubs or competitions.

The first half is a fast, hands-on tour of core techniques. Learners clean and explore real datasets with pandas, train classification models with scikit-learn and learn how to judge a model honestly using accuracy, confusion matrices and test sets. They then build a small neural network for image recognition and connect a language model to their own Python code through an API.

The second half runs like a tech team sprint. Learners form small teams, choose a real problem and dataset, split roles such as data lead, model lead and presenter, and build their project over three weeks with code reviews and check-ins. A dedicated responsible AI session asks teams to test their project for bias, privacy issues and failure cases before launch.

The bootcamp ends with Demo Day, where each team presents its problem, approach, results and lessons learned to classmates and families. The teacher gives detailed feedback on both the code and the presentation.

Every class uses Google Colab notebooks, so nothing needs to be installed and all work is saved for learners to keep building after the term.

How classes work

Each 75-minute session opens with a short concept talk and live-coded demo, followed by guided notebook exercises. From week seven, most of the time is team build time with teacher check-ins and code review.

What we'll cover

• 

  Week 1: How Machine Learning Works – Understand data, features, labels and models through a first quick experiment.

• 

  Week 2: Python and Data Wrangling with pandas – Clean, explore and visualize a real dataset.

• 

  Week 3: Classification Models with scikit-learn – Train and compare decision trees and other classifiers.

• 

  Week 4: Evaluating Models – Use accuracy, confusion matrices and test sets to spot overfitting and bias.

• 

  Week 5: Neural Networks and Image Recognition – Build a small neural network that classifies images.

• 

  Week 6: Working with Language Models and APIs – Connect a language model to Python code and design reliable prompts.

• 

  Week 7: Team Project Kickoff – Form teams, choose a problem and dataset, and assign roles.

• 

  Week 8: Build Sprint – Train, test and improve the team model with code reviews.

• 

  Week 9: Responsible AI Review and Testing – Check projects for bias, privacy and failure cases.

• 

  Week 10: Demo Day – Present the finished AI project and receive feedback.

Learners should be comfortable writing Python with variables, loops, functions and lists. Some algebra is helpful. No prior machine learning experience is required.