Cloud Code Agentic RAG Masterclass

Build an agentic RAG application from scratch by collaborating with Claude Code. Follow along with our video series using the docs in this repo.

Watch the full video on YouTube

What This Is

A hands-on course where you collaborate with Claude Code to build a full-featured RAG system. You're not the one writing code—Claude is. Your job is to guide it, understand what you're building, and course-correct when needed.

You don't need to know how to code. You do need to be technically minded and willing to learn about APIs, databases, and system architecture.

What You'll Build

• Chat interface with threaded conversations, streaming, tool calls, and subagent reasoning
• Document ingestion with drag-and-drop upload and processing status
• Full RAG pipeline: chunking, embedding, hybrid search, reranking
• Agentic patterns: text-to-SQL, web search, subagents with isolated context

Tech Stack

The 8 Modules

1. App Shell — Auth, chat UI, managed RAG with OpenAI Responses API
2. BYO Retrieval + Memory — Ingestion, pgvector, switch to generic completions API
3. Record Manager — Content hashing, deduplication
4. Metadata Extraction — LLM-extracted metadata, filtered retrieval
5. Multi-Format Support — PDF, DOCX, HTML, Markdown via Docling
6. Hybrid Search & Reranking — Keyword + vector search, RRF, reranking
7. Additional Tools — Text-to-SQL, web search fallback
8. Subagents — Isolated context, document analysis delegation

Getting Started

1. Clone this repo
2. Install Claude Code
3. Open in your IDE (Cursor, VS Code, etc.)
4. Run claude in the terminal
5. Use the /onboard command to get started

Docs

• PRD.md — What to build (the 8 modules in detail)
• CLAUDE.md — Context for Claude Code
• PROGRESS.md — Track your build progress

Join the Community

If you want to connect with hundreds of builders creating production-grade AI and RAG systems, join us in The AI Automators community. Share your progress, get help when you're stuck, and see what others are building.