How sundaranbu18/genai-course works
This repository combines Python teaching materials with two smaller RAG examples and a deployed Django RAG chatbot. In ragbot, users ask questions through the chat interface or embedded widget; Django routes requests to the RAG pipeline, which retrieves passages from a locally indexed document collection and asks Claude to answer from that context. The pipeline also supports conversation history. The graph includes the assignment RAG and single-file personal RAG as separate products. Some links rely on the README or SOURCE INDEX because their implementation excerpts were not provided.
Main entry points: pipeline.py RAG pipeline · views.py Django chat API · indexing.py Index builder · store.py FAISS retrieval
Components 22
Chat interface
- Embedded widget
ragbot/chatbot/static/chatbot/widget.js - Django chat API
ragbot/chatbot/views.py - Chat routes
ragbot/chatbot/urls.py
RAG runtime
- RAG pipeline
ragbot/ragcore/pipeline.py - Runtime settings
ragbot/ragcore/config.py - Text embeddings
ragbot/ragcore/embeddings.py - FAISS retrieval
ragbot/ragcore/store.py - Answer generation
ragbot/ragcore/generator.py - Conversation history
ragbot/ragcore/history.py
Document indexing
- Index builder
ragbot/ragcore/indexing.py - Document chunking
ragbot/ragcore/chunking.py - RAG CLI
ragbot/ragcore/cli.py
RAG examples
- Assignment API
rag_assisment/app.py - Assignment answer chain
rag_assisment/rag/chain.py - Assignment vector store
rag_assisment/rag/store.py - PDF ingestion
rag_assisment/rag/ingest.py - Evaluation questions
rag_assisment/evaluation/questions.py - Personal notes RAG
personal_rag/personal_rag.py
Learning materials
- Python course exercises
course
Other components
- Chat user
- Anthropic Claude
- Anthropic API
Connections 20
- Chat user → Django chat API: asks question
- Embedded widget → Django chat API: calls JSON API
- Chat routes → Django chat API: routes requests
ragbot/chatbot/urls.py - Django chat API → RAG pipeline: requests answer
ragbot/chatbot/views.py - RAG pipeline → Runtime settings: loads settings
ragbot/ragcore/pipeline.py - RAG pipeline → Conversation history: loads and records turns
ragbot/ragcore/pipeline.py - RAG pipeline → FAISS retrieval: searches passages
ragbot/ragcore/pipeline.py - RAG pipeline → Answer generation: generates answer
ragbot/ragcore/pipeline.py - FAISS retrieval → Text embeddings: embeds query
ragbot/ragcore/store.py - Answer generation → Anthropic API: requests completion
ragbot/ragcore/generator.py - RAG CLI → Index builder: builds index
ragbot/ragcore/cli.py - RAG CLI → RAG pipeline: answers questions
ragbot/ragcore/cli.py - Index builder → Document chunking: chunks documents
ragbot/ragcore/indexing.py - Index builder → FAISS retrieval: writes index
ragbot/ragcore/indexing.py - Assignment API → Assignment vector store: builds store
rag_assisment/app.py - Assignment API → Assignment answer chain: requests answer
rag_assisment/app.py - Assignment answer chain → Assignment vector store: retrieves passages
rag_assisment/rag/chain.py - Assignment vector store → PDF ingestion: loads chunks
rag_assisment/rag/store.py - Personal notes RAG → Chat user: answers questions
- Evaluation questions → Assignment answer chain: evaluates answers
rag_assisment/assigment.py