Lets Build an Agent for Schools (For the future sessions)

• Basic Idea:
  • Make AI agent prepare question papers
  • Also make it analyze the results and guide student with improvement areas and references to work on

Lets Build a RAG for Students

• Basic Idea:
  • RAG which is based on standard text books of boards
  • A student can ask questions and this RAG should help answering questions of student by giving references from books
  • Convert that into simpler language using Generator

10,000 feet approach for solving

Phase 1 : Insert documents as vectors in Vector DB

• Overview
  

• How to parse pdf files and pull out text and image sections

• How do i chunk ?
• Which Embedding for Text and Vector ?
• Which Vector Database

Phase 2: Using Vector DB as Retriever

• Test the retrieval
• Understand vector DB Retrievals

Phase 3: Building RAG (Version 1)

• Version 1 will be just trying to find from Vectors, if not available will not answer.
• Explore Frameworks:
  • langchain
• Add Generator
  • with local llm
  • with openai/gemini

What skills are required ?

• Python
  • Core
  • external libraries
    • Data manipulation
    • pdf extractions
• Embeddings
  • Text
  • Image
• Cloud (optional)
• RAG Frameworks
  • langchain
• Vector Databases
  • Pinecone
  • Weaviate/Qdrant/Milvus
• LLM:
  • OpenAI
  • Gemini
  • Anthropic
  • LLAMA
• SLM:
  • Phi
• Transformers:
  • Pytorch/Tensorflow
  • Hugging face
• Frontend:
  • Streamlit
  • FastAPI
• NLP
• Docker
• Git
• Deployments

Goal:

• Git, Visual studio code usage
• Python:
  • core
  • external libraries