Level 3Usually Grades 5–6 · Ages 10–12 · 60-minute sessions
Python & Their First AI Application
This is the bridge to real, professional programming. Your child moves from drag-and-drop blocks to writing actual Python, the language used by engineers at companies like Google and OpenAI. It is also the year your child stops chatting with AI and starts programming it.
What your child will learn
- Write real Python programs with variables, loops, functions, lists, and dictionaries
- Work with real data, build charts, and train a machine-learning classifier with scikit-learn
- Measure how accurate a model really is, and learn why a high score can still hide a problem
- Call an AI language model from inside their own Python code and use the result in a program
- Practice prompt engineering as an engineering skill, with clear instructions, examples, and tests
- Use APIs to pull real-world data, and use Git and GitHub the way professionals do
What your child will build
- Their first complete, end-to-end machine-learning project
- A Python application powered by a real AI model, such as a study helper or a story generator
- A personal portfolio website
- A capstone project of their choosing, saved to a class portfolio (published on GitHub once your child is 13, with a parent's OK)
Tools
Python (VS Code and Replit)pandasmatplotlibscikit-learnJupyter and Google ColabAn AI model API through the iLearnAI sandboxGitHub
By year end: Your child writes Python from a blank file, trains a machine-learning model, and has built a working application that calls a real AI model and handles its answers responsibly.
Level 4Usually Grades 7–8 · Ages 12–14 · 60-minute sessions
Full-Scale Programming, Applied AI & First Agents
Your child is treated as an apprentice developer, building full-scale programs and real machine-learning applications while practicing the habits of professional software engineers. This is the year your child builds their first AI agent: a program that reasons, chooses a tool, uses it, and acts.
What your child will learn
- Write structured programs using object-oriented design and core computer-science algorithms
- Build complete machine-learning pipelines, including data cleaning, training, and error analysis
- Create computer-vision and language applications such as face detection, sentiment analysis, and spam filters
- Build a retrieval system so an AI assistant can answer questions using their own documents and cite where the answer came from
- Give an AI model real tools to call, such as a calculator, a search function, or a file reader, and watch it reason through a multi-step task
- Deploy a finished application to the web, and use Git, code review, and testing like a professional team
What your child will build
- An end-to-end machine-learning web application with a trained model behind it
- An AI assistant that answers questions from a document set your child chooses
- A tool-using AI agent that completes a real multi-step task
- A capstone project with a written evaluation of how well it performs
Tools
PythonJavapandasscikit-learnOpenCVTensorFlow and KerasFlask and FastAPIVector searchAI model APIs with tool callingGitHub
By year end: Your child has shipped a working AI application to the web, understands how an AI agent decides what to do next, and keeps a GitHub repository with real commits, tests, and documentation.
