# ankit0514/rag_project architecture

> Architecture diagram of the GitHub repository ankit0514/rag_project, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-20). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/ankit0514/rag_project
- Repository: https://github.com/ankit0514/rag_project

## Overview

This is an end-to-end document RAG demo: a user uploads a PDF, DOCX, or TXT, then asks a question and receives an answer with source snippets. Uploads are loaded, chunked, embedded, and indexed; chat combines vector and BM25 retrieval, reranks results, builds a prompt, and generates an answer using Gemini or a local fallback. The graph also includes conversation memory and the repository’s database models. Some subsystems are represented by paths rather than inspected implementation details; edges are limited to relationships shown in excerpts, the source index, or README.

Main entry points: [`backend/app/services/chat_service.py`](https://github.com/ankit0514/rag_project/blob/main/backend/app/services/chat_service.py) (Chat service), [`backend/app/retrieval/vector_search.py`](https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/vector_search.py) (Vector search), [`backend/app/retrieval/hybrid_search.py`](https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/hybrid_search.py) (Hybrid search), [`backend/app/database/session.py`](https://github.com/ankit0514/rag_project/blob/main/backend/app/database/session.py) (Database session)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_api["API and interface"]
  node_frontend["Demo frontend<br/>[index.html]"]
  node_app["FastAPI app<br/>[main.py]"]
  node_rag_api["RAG routes<br/>[rag.py]"]
end

subgraph group_ingest["Document ingestion"]
  node_ingest_pipeline["Format loaders<br/>[pipeline.py]"]
  node_loaders["PDF DOCX TXT loaders"]
  node_chunker["Text chunking"]
  node_embeddings["Embedding service"]
end

subgraph group_retrieval["Search and ranking"]
  node_vector_search["Vector search<br/>[vector_search.py]"]
  node_faiss[("FAISS index<br/>[faiss_store.py]")]
  node_hybrid["Hybrid search<br/>[hybrid_search.py]"]
  node_bm25["BM25 retrieval<br/>[bm25.py]"]
  node_reranker["Result reranker<br/>[reranker.py]"]
end

subgraph group_answer["Chat and generation"]
  node_chat_service["Chat service<br/>[chat_service.py]"]
  node_prompt["Prompt builder<br/>[prompt_builder.py]"]
  node_prompts["Prompt templates<br/>[prompts.py]"]
  node_gemini["Gemini and fallback<br/>[gemini_client.py]"]
  node_memory[("Conversation memory")]
end

subgraph group_state["State and configuration"]
  node_config["Settings<br/>[config.py]"]
  node_db_session[("Database session<br/>[session.py]")]
  node_chat_model["Chat history model<br/>[chat.py]"]
  node_document_model["Document model<br/>[document.py]"]
  node_user_model["User model<br/>[user.py]"]
end

node_user(("User"))

node_user -->|"uses"| node_frontend
node_frontend -->|"opens demo"| node_app
node_app -->|"registers routes"| node_rag_api
node_user -->|"uploads and asks"| node_rag_api
node_rag_api -->|"delegates"| node_chat_service
node_chat_service -->|"ingests and searches"| node_hybrid
node_ingest_pipeline -->|"dispatches by format"| node_loaders
node_vector_search -->|"loads document"| node_ingest_pipeline
node_vector_search -->|"chunks text"| node_chunker
node_vector_search -->|"embeds text"| node_embeddings
node_vector_search -->|"indexes and searches"| node_faiss
node_hybrid -->|"gets vector results"| node_vector_search
node_hybrid -->|"gets keyword results"| node_bm25
node_hybrid -->|"reranks combined results"| node_reranker
node_chat_service -->|"builds prompt"| node_prompt
node_prompt -->|"uses templates"| node_prompts
node_chat_service -->|"generates answer"| node_gemini
node_chat_service -->|"reads and updates"| node_memory
node_db_session -->|"reads settings"| node_config
node_chat_model -->|"imports base"| node_db_session
node_document_model -->|"imports base"| node_db_session
node_user_model -->|"imports base"| node_db_session

click node_frontend "https://github.com/ankit0514/rag_project/blob/main/frontend/index.html"
click node_app "https://github.com/ankit0514/rag_project/blob/main/backend/app/main.py"
click node_rag_api "https://github.com/ankit0514/rag_project/blob/main/backend/app/api/routes/rag.py"
click node_chat_service "https://github.com/ankit0514/rag_project/blob/main/backend/app/services/chat_service.py"
click node_ingest_pipeline "https://github.com/ankit0514/rag_project/blob/main/backend/app/ingestion/pipeline.py"
click node_loaders "https://github.com/ankit0514/rag_project/tree/main/backend/app/ingestion/loaders"
click node_chunker "https://github.com/ankit0514/rag_project/blob/main/backend/app/ingestion/chunking/recursive_chunker.py"
click node_vector_search "https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/vector_search.py"
click node_embeddings "https://github.com/ankit0514/rag_project/blob/main/backend/app/embeddings/embedding_service.py"
click node_faiss "https://github.com/ankit0514/rag_project/blob/main/backend/app/vectorstore/faiss_store.py"
click node_hybrid "https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/hybrid_search.py"
click node_bm25 "https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/bm25.py"
click node_reranker "https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/reranker.py"
click node_prompt "https://github.com/ankit0514/rag_project/blob/main/backend/app/llm/prompt_builder.py"
click node_prompts "https://github.com/ankit0514/rag_project/blob/main/backend/app/llm/prompts.py"
click node_gemini "https://github.com/ankit0514/rag_project/blob/main/backend/app/llm/gemini_client.py"
click node_memory "https://github.com/ankit0514/rag_project/blob/main/backend/app/memory/conversation_memory.py"
click node_config "https://github.com/ankit0514/rag_project/blob/main/backend/app/core/config.py"
click node_db_session "https://github.com/ankit0514/rag_project/blob/main/backend/app/database/session.py"
click node_chat_model "https://github.com/ankit0514/rag_project/blob/main/backend/app/database/models/chat.py"
click node_document_model "https://github.com/ankit0514/rag_project/blob/main/backend/app/database/models/document.py"
click node_user_model "https://github.com/ankit0514/rag_project/blob/main/backend/app/database/models/user.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_frontend,node_app,node_rag_api,node_user toneBlue
class node_ingest_pipeline,node_loaders,node_chunker,node_embeddings toneAmber
class node_vector_search,node_faiss,node_hybrid,node_bm25,node_reranker toneMint
class node_chat_service,node_prompt,node_prompts,node_gemini,node_memory toneRose
class node_config,node_db_session,node_chat_model,node_document_model,node_user_model toneIndigo
```

## Components

### API and interface

- [Demo frontend](https://github.com/ankit0514/rag_project/blob/main/frontend/index.html): `frontend/index.html`
- [FastAPI app](https://github.com/ankit0514/rag_project/blob/main/backend/app/main.py): `backend/app/main.py`
- [RAG routes](https://github.com/ankit0514/rag_project/blob/main/backend/app/api/routes/rag.py): `backend/app/api/routes/rag.py`

### Document ingestion

- [Format loaders](https://github.com/ankit0514/rag_project/blob/main/backend/app/ingestion/pipeline.py): `backend/app/ingestion/pipeline.py`
- [PDF DOCX TXT loaders](https://github.com/ankit0514/rag_project/tree/main/backend/app/ingestion/loaders): `backend/app/ingestion/loaders`
- [Text chunking](https://github.com/ankit0514/rag_project/blob/main/backend/app/ingestion/chunking/recursive_chunker.py): `backend/app/ingestion/chunking/recursive_chunker.py`
- [Embedding service](https://github.com/ankit0514/rag_project/blob/main/backend/app/embeddings/embedding_service.py): `backend/app/embeddings/embedding_service.py`

### Search and ranking

- [Vector search](https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/vector_search.py): `backend/app/retrieval/vector_search.py`
- [FAISS index](https://github.com/ankit0514/rag_project/blob/main/backend/app/vectorstore/faiss_store.py): `backend/app/vectorstore/faiss_store.py`
- [Hybrid search](https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/hybrid_search.py): `backend/app/retrieval/hybrid_search.py`
- [BM25 retrieval](https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/bm25.py): `backend/app/retrieval/bm25.py`
- [Result reranker](https://github.com/ankit0514/rag_project/blob/main/backend/app/retrieval/reranker.py): `backend/app/retrieval/reranker.py`

### Chat and generation

- [Chat service](https://github.com/ankit0514/rag_project/blob/main/backend/app/services/chat_service.py): `backend/app/services/chat_service.py`
- [Prompt builder](https://github.com/ankit0514/rag_project/blob/main/backend/app/llm/prompt_builder.py): `backend/app/llm/prompt_builder.py`
- [Prompt templates](https://github.com/ankit0514/rag_project/blob/main/backend/app/llm/prompts.py): `backend/app/llm/prompts.py`
- [Gemini and fallback](https://github.com/ankit0514/rag_project/blob/main/backend/app/llm/gemini_client.py): `backend/app/llm/gemini_client.py`
- [Conversation memory](https://github.com/ankit0514/rag_project/blob/main/backend/app/memory/conversation_memory.py): `backend/app/memory/conversation_memory.py`

### State and configuration

- [Settings](https://github.com/ankit0514/rag_project/blob/main/backend/app/core/config.py): `backend/app/core/config.py`
- [Database session](https://github.com/ankit0514/rag_project/blob/main/backend/app/database/session.py): `backend/app/database/session.py`
- [Chat history model](https://github.com/ankit0514/rag_project/blob/main/backend/app/database/models/chat.py): `backend/app/database/models/chat.py`
- [Document model](https://github.com/ankit0514/rag_project/blob/main/backend/app/database/models/document.py): `backend/app/database/models/document.py`
- [User model](https://github.com/ankit0514/rag_project/blob/main/backend/app/database/models/user.py): `backend/app/database/models/user.py`

### Other components

- **User**

## Connections

- User → Demo frontend: uses (evidence: `README.md`)
- Demo frontend → FastAPI app: opens demo (evidence: `README.md`)
- FastAPI app → RAG routes: registers routes (evidence: `backend/app/main.py`)
- User → RAG routes: uploads and asks (evidence: `README.md`)
- RAG routes → Chat service: delegates (evidence: `backend/app/api/routes/rag.py`)
- Chat service → Hybrid search: ingests and searches (evidence: `backend/app/services/chat_service.py`)
- Format loaders → PDF DOCX TXT loaders: dispatches by format (evidence: `backend/app/ingestion/pipeline.py`)
- Vector search → Format loaders: loads document (evidence: `backend/app/retrieval/vector_search.py`)
- Vector search → Text chunking: chunks text (evidence: `backend/app/retrieval/vector_search.py`)
- Vector search → Embedding service: embeds text (evidence: `backend/app/retrieval/vector_search.py`)
- Vector search → FAISS index: indexes and searches (evidence: `backend/app/retrieval/vector_search.py`)
- Hybrid search → Vector search: gets vector results (evidence: `backend/app/retrieval/hybrid_search.py`)
- Hybrid search → BM25 retrieval: gets keyword results (evidence: `backend/app/retrieval/hybrid_search.py`)
- Hybrid search → Result reranker: reranks combined results (evidence: `backend/app/retrieval/hybrid_search.py`)
- Chat service → Prompt builder: builds prompt (evidence: `backend/app/services/chat_service.py`)
- Prompt builder → Prompt templates: uses templates (evidence: `backend/app/llm/prompt_builder.py`)
- Chat service → Gemini and fallback: generates answer (evidence: `backend/app/services/chat_service.py`)
- Chat service → Conversation memory: reads and updates (evidence: `backend/app/services/chat_service.py`)
- Database session → Settings: reads settings (evidence: `backend/app/database/session.py`)
- Chat history model → Database session: imports base (evidence: `backend/app/database/models/chat.py`)
- Document model → Database session: imports base (evidence: `backend/app/database/models/document.py`)
- User model → Database session: imports base (evidence: `backend/app/database/models/user.py`)
