Hands-on

Projects

Builds

About These Projects

Watching lessons only gets you so far. These five builds take the ideas from the modules and make you apply them: you'll build a chatbot with memory, pull clean data out of free text with structured outputs, answer questions from your own documents with RAG, build an agent that uses tools, and finish by shipping a full-stack AI app.

Each project follows along with a video walkthrough. Work through them in order, or jump to the one that matches the module you just finished.

Projects

5 builds
01

Build a Chatbot with Memory

Connect a model to a simple chat interface and keep the conversation history so it can answer follow-up questions.

02

Extract Structured Data from Text

Turn unstructured text into validated data by defining a Pydantic schema and passing it to the model as the response format.

03

Build a RAG Question-Answering Tool for PDFs

Load a PDF, split and embed it into a vector store, and answer questions from its contents with retrieval and a model.

04

Build an AI Agent from Scratch

Write the agent loop yourself with function calling, add RAG as one of its tools, and watch its reasoning trace in a live app.

05

Build and Deploy a Full-Stack AI App

Put everything together by building an AI-powered email client with Next.js and deploying it live on Vercel.