nandishh18/rag-agent-judge

How nandishh18/rag-agent-judge works

This repository is a paper-focused RAG chatbot. Its main workflow builds a local searchable index from a PDF, then embeds a user question, retrieves relevant chunks, and sends them with the question to Groq for a grounded answer. It also supports generating a Markdown answer set for 15 assignment questions. The relationships below are supported by the sampled source and README; the generated index is documented but is not present in the file tree.

Main entry points: rag_chat.py Chunk Retrieval · build_index.py Index Builder · answer_questions.py Question Runner · ingest.py PDF Extraction

Components 16

Indexing

Index Builder src/build_index.py
PDF Extraction src/ingest.py
Text Embeddings src/build_index.py
FAISS Index src/build_index.py
Chunk Metadata src/build_index.py

Question Answering

Interactive Chat src/rag_chat.py
RAG Chatbot src/rag_chat.py
Query Embedding src/rag_chat.py
Chunk Retrieval src/rag_chat.py
Prompt Assembly src/rag_chat.py
Grounded Answer src/rag_chat.py

Batch Answers

Question Runner src/answer_questions.py
Answer Report src/answer_questions.py

Other components

Paper PDF data/agent_as_a_judge.pdf
User
Groq LLM

Connections 17

Paper PDF → PDF Extraction: extracts pages README.md
PDF Extraction → Index Builder: loads chunks src/build_index.py
Index Builder → Text Embeddings: embeds chunks src/build_index.py
Text Embeddings → FAISS Index: adds vectors src/build_index.py
Index Builder → Chunk Metadata: writes metadata src/build_index.py
User → Interactive Chat: asks questions README.md
Interactive Chat → RAG Chatbot: calls answer src/rag_chat.py
RAG Chatbot → Chunk Retrieval: retrieves chunks src/rag_chat.py
Chunk Retrieval → Query Embedding: embeds query src/rag_chat.py
Chunk Retrieval → FAISS Index: searches index src/rag_chat.py
Chunk Retrieval → Chunk Metadata: reads chunks src/rag_chat.py
RAG Chatbot → Prompt Assembly: assembles context src/rag_chat.py
Prompt Assembly → Groq LLM: requests completion src/rag_chat.py
Groq LLM → Grounded Answer: returns answer src/rag_chat.py
Grounded Answer → Interactive Chat: prints answer src/rag_chat.py
Question Runner → RAG Chatbot: calls answer src/answer_questions.py
Question Runner → Answer Report: writes report src/answer_questions.py

More Python diagrams: intellindust-ai-lab/deim · nvlabs/fast-foundationstereo · gramaddict/bot · shisatoyano/autonomousvehiclecontrolbeginnersguide · pypose/pypose

Drawn by GitDiagram on September 30, 2026 from the repository's file tree, README and source files. Read as Markdown ·