# srishhhhx/multidoc-adaptive-rag architecture

> Architecture diagram of the GitHub repository srishhhhx/multidoc-adaptive-rag, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-10-10). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/srishhhhx/multidoc-adaptive-rag
- Repository: https://github.com/srishhhhx/multidoc-adaptive-rag

## Overview

##### Project type and purpose
- This is a full-stack AI application (“Multi-Doc Adaptive RAG Agent”) that combines a React.js frontend with a Python/FastAPI backend.
- Its goal is to answer complex natural-language queries by orchestrating a Retrieval-Augmented Generation (RAG) pipeline across multiple documents and real-time web search, with self-correcting loops and quality gates.

##### File structure
- Top-level components:
- api.py, config.py, session_manager.py, state.py, utils.py, rag_workflow.py, multimodal_loader.py, document_loader.py, document_processor.py — these form the FastAPI backend and LangGraph workflow definition.
- chains/ – modular LangChain/LangGraph “nodes” for each pipeline step (query router, reranker, generator, evaluators, etc.).
- frontend/ – React/Vite/Tailwind-based UI, Dockerfile, nginx.conf for production.
- Key patterns:
- A layered architecture: UI → HTTP API → Orchestrator (rag_workflow.py) → chain nodes → external services/data stores.
- A dynamic graph pipeline in LangGraph, with quality‐gate loops (self-correction) and a router that chooses tools.
- Configuration & deployment:
- .env.example for keys, Dockerfile for frontend, nginx.conf, Uvicorn for backend.

##### README for additional insights
- The README confirms the tech stack (React, FastAPI, LangGraph, FAISS/Chroma, Tavily, Gemini LLMs).
- It describes performance optimizations (batch LLM calls, GPU reranking).
- It outlines advanced pipeline features: Query Analysis Router, cross-encoder reranking, self-correction loops, circuit breakers.

## Diagram (Mermaid)

```mermaid
flowchart TB
    %% Frontend Layer
    subgraph "Frontend UI"
        direction TB
        User["User"]:::external
        FrontendDir["frontend/"]:::ui
        ReactSrc["frontend/src/"]:::ui
        NginxConf["nginx.conf"]:::ui
        FrontendDocker["Dockerfile"]:::ui
    end

    %% Backend Layer
    subgraph "Backend API"
        direction TB
        API["api.py"]:::backend
        Config["config.py"]:::backend
        SessionMgr["session_manager.py"]:::backend
        State["state.py"]:::backend
        Utils["utils.py"]:::backend
        Requirements["requirements.txt"]:::backend
    end

    %% Orchestration Engine
    Orchestrator["rag_workflow.py (LangGraph)"]:::backend

    %% Pipeline Nodes
    subgraph "Pipeline (LangGraph Nodes)"
        direction TB
        Router["Query Analysis Router"]:::pipeline
        Retriever["Document Retrieval & Web Search"]:::pipeline
        Reranker["Rerank Documents"]:::pipeline
        Generator["Generate Answer"]:::pipeline
        Evaluator["Quality Gates (evaluate)"]:::pipeline
        ContextAssess["Context Assessment"]:::pipeline
        RewriteQuery["Rewrite Query"]:::pipeline
    end

    %% External Services & Data Stores
    subgraph "External Services & Data Stores"
        direction TB
        VectorDB["Vector Database (FAISS/ChromaDB)"]:::external
        WebSearch["Web Search API (Tavily)"]:::external
        LLMAPI["LLM Providers (Gemini)"]:::external
        FileStore["File Store (uploaded docs)"]:::external
    end

    %% Connections
    User -->|"enters query"| ReactSrc
    ReactSrc -->|"Axios request"| API
    API -->|invokes| Orchestrator
    Orchestrator -->|routes to| Router
    Router -->|calls| Retriever
    Retriever -->|retrieved chunks| Reranker
    Reranker -->|ranked chunks| Generator
    Generator -->|candidate answers| Evaluator
    Evaluator -->|checks pass| API
    API -->|"response"| ReactSrc

    %% Self-correction Loop
    Evaluator -- "fail checks" --> ContextAssess
    ContextAssess --> RewriteQuery
    RewriteQuery --> Router

    %% External calls
    Retriever -->|vector query| VectorDB
    Retriever -->|web query| WebSearch
    Generator -->|LLM call| LLMAPI
    Retriever -->|load docs| FileStore

    %% Click Events
    click FrontendDir "https://github.com/srishhhhx/multidoc-adaptive-rag/tree/main/frontend/"
    click ReactSrc "https://github.com/srishhhhx/multidoc-adaptive-rag/tree/main/frontend/src/"
    click NginxConf "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/frontend/nginx.conf"
    click FrontendDocker "https://github.com/srishhhhx/multidoc-adaptive-rag/tree/main/frontend/Dockerfile"
    click API "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/api.py"
    click Config "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/config.py"
    click SessionMgr "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/session_manager.py"
    click State "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/state.py"
    click Utils "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/utils.py"
    click Requirements "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/requirements.txt"
    click Orchestrator "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/rag_workflow.py"
    click Router "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/chains/query_analysis_router.py"
    click Retriever "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/multimodal_loader.py"
    click Reranker "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/chains/rerank_documents.py"
    click ContextAssess "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/chains/context_assessment.py"
    click RewriteQuery "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/chains/rewrite_query.py"
    click Generator "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/chains/generate_answer.py"
    click Evaluator "https://github.com/srishhhhx/multidoc-adaptive-rag/blob/main/chains/evaluate.py"

    %% Styles
    classDef ui fill:#D0E8FF,stroke:#000,stroke-width:1px
    classDef backend fill:#D0FFD6,stroke:#000,stroke-width:1px
    classDef pipeline fill:#FFE0B2,stroke:#000,stroke-width:1px
    classDef external fill:#E1BEE7,stroke:#000,stroke-width:1px
```
