DSG IIT Roorkee Bootcamp
  • Start Here: Welcome to the Bootcamp
    • Course Structure
    • Course Syllabus and Timelines
    • Know your Educators
    • Action Items and Prerequisites
    • Kick-Off Session for the Bootcamp
  • Module 1: Basics of LLMs
    • What is Generative AI?
    • What is a Large Language Model?
    • Advantages and Applications of LLMs
    • Bonus Resource: Multimodal LLMs and Google Gemini
  • Module 2: Word Vectors, Simplified
    • What is a Word Vector?
    • Word Vector Relationships
    • Role of Context in LLMs
    • Transforming Vectors into LLM Responses
    • Bonus: Overview of the Transformer Architecture
      • Attention Mechanism
      • Multi-Head Attention and Transformer Architecture
      • Vision Transformers (ViTs)
    • Bonus: Future of LLMs? | By Transformer Co-inventor
    • Graded Quiz 1
  • Module 3: Prompt Engineering and Token Limits
    • What is Prompt Engineering
    • Prompt Engineering and In-context Learning
    • For Starters: Best Practices
    • Navigating Token Limits
    • Hallucinations in LLMs
    • Prompt Engineering Excercise (Ungraded)
      • Story for the Excercise: The eSports Enigma
      • Your Task fror the Module
  • Module 4: RAG and LLM Architecture
    • What is Retrieval Augmented Generation (RAG)?
    • Primer to RAG: Pre-trained and Fine-Tuned LLMs
    • In-context Learning
    • High-level LLM Architecture Components for In-context Learning
    • Diving Deeper: LLM Architecture Components
    • Basic RAG Architecture with Key Components
    • RAG versus Fine-Tuning and Prompt Engineering
    • Versatility and Efficiency in RAG
    • Key Benefits of using RAG in an Enterprise/Production Setup
    • Hands-on Demo: Performing Similarity Search in Vectors (Bonus Module)
      • Implementation of RAG for Production Use Cases
    • Using kNN and LSH to Enhance Similarity Search (Bonus Module)
    • Bonus Video: Implementing End-to-End RAG | 1-Hour Session
    • Graded Quiz 2
  • Module 5: Hands-on Development
    • Build Your Own AI Slide Search Pipeline
    • Prerequisites (Must)
    • Docker Basics
    • Your Hands-on RAG Journey
    • 1 – First RAG Pipeline
      • Building with Open AI
      • How it Works
      • RAG with Gemini and other Open AI Alternatives
      • RAG with Open Source and Running "Examples"
    • 2 – Amazon Discounts App
      • How the Project Works
      • Building the App
    • 3 – Private RAG with Mistral, Ollama and Pathway
      • Building a Private RAG project
      • (Bonus) Adaptive RAG Overview
    • 4 – Realtime RAG with LlamaIndex/LangChain and Pathway
      • Understand the Basics
      • Implementation with LlamaIndex and LangChain
    • Cloud Deployment Basics
  • Module 6: Project Tracks and Submission
    • Prizes and Giveaways
    • Suggested Tracks for Ideation
    • Sample Projects and Additional Resources
    • Submit Project for Review
Powered by GitBook
On this page
  • GitHub Repository
  • Features of the AI Slide Search App
  • Architecture Overview
  1. Module 5: Hands-on Development

Build Your Own AI Slide Search Pipeline

PreviousModule 5: Hands-on DevelopmentNextPrerequisites (Must)

Last updated 7 months ago

By the end of this bootcamp, you will be equipped to create, deploy, and manage complex AI pipelines just like the demo available here: đź”—

Here's a short video showcasing the product demo:

Pathway Slide Search solution has been showcased on Intel Tiber Cloud, during the .

GitHub Repository

Features of the AI Slide Search App

💡 Instant Search: Retrieve relevant slides in seconds. ⚡ Real-Time Updates: Indexing happens instantly when files are added, removed, or modified. 📂 Supports Multiple Data Sources: Connect with local folders, SharePoint, Google Drive, and more. 🔍 Advanced Metadata Extraction: Utilize Pathway's vision-language models to process PDF and PowerPoint slides. 🛠️ Flexible and Customizable: Modify schema to fit your needs.

Architecture Overview

This app template uses Pathway to:

  1. Ingest and Parse Data: Extract content from PDFs and PPTs using Pathway’s SlideParser.

  2. Generate Embeddings: Use OpenAI’s text-embedding-ada-002 or a local embedding model to index slide content.

  3. Store Embeddings in a Vector Store: Store indexed data locally for fast retrieval.

  4. Serve a UI for Search Queries: Use Streamlit UI to interact with the search pipeline.

Next Steps:

  1. Complete the Hands-On Development Module to learn how to create these AI pipelines

  2. Explore Integrating RAG Pipelines with Local LLMs for cost-effective and private AI solutions.

đź“‚ You will use this repository to set up and run the demo locally or on the cloud. The repo provides all the necessary code and configurations to get started.

Slides AI Search Repository
Slides AI Search Demo
Intel AI Summit at Paris