How AI “Reads” Documents: Design a RAG Pipeline 🤖

Generative AI is transforming how we interact with data, but how do these systems actually work under the hood?

In this interactive session, we will tackle one of the most common architectures in modern AI: Retrieval-Augmented Generation (RAG).
We will design the backend infrastructure that allows a chatbot to "read" private documents (like PDFs or company wikis) and accurately answer user questions based on that text.
We’ll walk through how PDFs and text files are processed, chunked, transformed into embeddings, and stored for fast retrieval.

Agenda (90 Minutes):

• 15 mins: Introductions and scoping the requirements (identifying scale, latency, and constraints).
• 40 mins: High-Level Design (drawing the happy path from client to database and back).
• 20 mins: Deep Dive (exploring bottlenecks, edge cases, and specific database choices).
• 15 mins: Wrap-up, Q&A, and networking.

Perfect for anyone curious about system design, backend engineering, or AI systems.
*No prior experience with RAG required*