Builder path

Guided course for want to build with ai. 8 lessons, about 6 h 45 min.

For students who want to build with AI: code, grounded assistants, agents and automations. You'll need some comfort with a computer and ideally a little code, but each lesson explains what it assumes. The tutor can help whenever a concept is new.

By the end, you'll be able to

• Explain how LLMs work well enough to design around their limits.
• Write prompts for systems, not just chats.
• Use AI coding tools and review what they produce.
• Build a grounded assistant (RAG) and an agent with tools.
• Ship a small automation and evaluate whether it works.

Unit 1: Under the hood

Builders need a precise model of the model: tokens, context, reasoning, and how prompts behave when they're part of a system rather than a chat.

• How large language models work: Tokens, transformers, training and why context windows matter.
• Prompting that works: Context, examples, constraints and iteration.
• Reasoning models: Why some models “think” before answering, and when that's worth the wait.

Unit 2: Building with models

Now build: code with AI, ground a model in your own data, and give it tools. Each lesson ends with something that runs.

• Coding with AI: From autocomplete to agents, and how to stay the engineer.
• RAG: AI over your own documents: How tools answer from your files instead of from memory.
• AI agents: Models that take actions: browsing, running code and using your apps.

Unit 3: Shipping it

Automate a real workflow and work with images, audio and video, then evaluate what you made before others rely on it.

• Automating routine work: Connect apps and add AI steps to workflows you repeat every week.
• Images, video and voice: What AI can see, hear and generate beyond text.

Capstone project: Ship a small, evaluated AI tool

Build a small tool that solves a real problem for you or a club, such as a grounded Q&A over your notes or an automation that saves a weekly chore, and evaluate it honestly.