Build a "Chat With Your Docs" App
Turn Your Documents Into a Sourced, Question-Answering Assistant
Turn a pile of documents into something you can simply ask. In about an hour, you'll build a working "Chat With Your Docs" assistant (a retrieval-augmented generation (RAG) app over your own files) that answers plain-English questions with accurate, sourced answers instead of making you dig through PDFs. It's a hands-on build on current Claude tooling and a lightweight retrieval setup, deliberately a build and not an ML theory course, with the concepts taught just-in-time.
with GenAI Academy
GenAI Academy
Course Overview
Stop digging through PDFs. This short, hands-on course turns your own documents into an assistant you can simply ask - type a plain-English question, get an accurate answer back, with a citation pointing to the exact file it came from.
You'll build a real retrieval-augmented generation (RAG) app over your files using Claude and a lightweight retrieval setup: chunk and embed your documents, store them in a simple local vector store, retrieve the right context for each question, and wire Claude to answer grounded only in what it found. It's deliberately a build, not an ML theory course - the concepts are taught just-in-time, as you need them.
By the end you'll have one working assistant over your own files, the skill to harden it so it says "I don't know" rather than hallucinate, and a clear sense of how to extend it - all in about an hour, no machine-learning background required.
What you'll learn
Explain what RAG is and why it beats pasting documents into a chat window — accuracy, sources, and scale
Build a document ingestion pipeline that chunks and embeds your own files into a simple local vector store
Implement retrieval that surfaces the most relevant passages for any given question
Wire Claude to answer questions grounded only in retrieved context, returning sourced answers that cite the file they came from
Evaluate and harden the assistant so it says "I don't know" instead of hallucinating, and test it on real questions
Run a complete, working "Chat With Your Docs" assistant over your own files and identify how to extend it
Meet the instructors
GenAI Academy
GenAI Academy
GenAI Academy builds practical, hands-on AI courses designed so you apply what you learn the same day. This course is produced by the Academy's in-house team to the same standard as our instructor-led courses.
Who this course is for
Analysts & Knowledge Workers Who Code a Little Ops and knowledge workers comfortable running a terminal command and editing a config file who want a sourced answer engine over their own documents
Technical PMs & Solo Builders Product people and independent builders who want to ship a genuinely useful RAG app fast, without standing up heavy infrastructure
Junior Developers Early-career devs who want to learn RAG by building a real, working assistant — not by wading through ML theory
What's included
7 self-paced video lessons (build a working RAG assistant end to end)
A tested starter repo / scaffold with a /docs folder, .env.example, and pinned dependencies
A completed reference repo — the final working app for comparison
Copy-paste code snippets for every module
A one-page "RAG in plain English" cheat sheet
A sample document set so you can build even without your own files
Course syllabus
7 lessons
Why "Chat With Your Docs" (and What RAG Actually Is)
Setup: Project, API Key, and Your Files
Make Your Docs Searchable (Chunk + Embed + Store)
Retrieve the Right Context for a Question
Answer With Sources (Wire Claude + Grounding Prompt)
Make It Usable + Trustworthy
Ship It & Extend (Capstone Wrap)
Schedule
Format
Self-paced video course
Lessons
7 lessons
Price
$15
Level
Intermediate
Coming Soon
Coming Soon
Have questions about whether this course is right for you? Reach out to us.
Contact usRecommended For You
Create a free account
Sign up to unlock personalized course recommendations, track your progress, and save your favorites.
Sign Up Free
Already have an account? Log in