# itanishqshelar/smartrag architecture

> Architecture diagram of the GitHub repository itanishqshelar/smartrag, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-10-30). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/itanishqshelar/smartrag
- Repository: https://github.com/itanishqshelar/smartrag

## Overview

##### Project type and purpose
- Full-stack, locally-hosted multimodal RAG (Retrieval-Augmented Generation) system with a Streamlit “chat” front-end.
- Allows users to ingest documents, images, and audio, stores metadata in SQLite, embeddings in ChromaDB/FAISS, and uses a local LLM (Llama 3.1 via Ollama) for responses.

##### File structure
- Top-level scripts:
- chatbot_app.py (Streamlit UI entry point)
- start.py / clean_db.py (utility scripts)
- Configuration: config.yaml + config_schema.py for Pydantic validation
- multimodal_rag/: core library
- base.py (interfaces)
- system.py (orchestrator)
- processors/ (document, image, audio extraction & chunking)
- vector_stores/ (ChromaDB, FAISS adapters)
- Persistence: file_storage.db (SQLite), vector_db/ or faiss_db/ directories
- docker/: Dockerfiles and compose for containerized deployment

##### Extract insights from README
- Processing pipeline: extract → chunk → embed → store
- Query pipeline: embed query → semantic search → retrieve contexts → LLM inference → response
- Tech stack: Python, Streamlit, SQLite, Ollama, PyTorch, Transformers, Whisper, BLIP, Tesseract, ChromaDB, FAISS

## Diagram (Mermaid)

```mermaid
flowchart TD
    %% Terminators
    Start(("Start")):::terminator
    End(("End")):::terminator

    %% Frontend Layer
    subgraph "Frontend Layer"
      UI["Streamlit UI<br/>(Streamlit)"]:::frontend
    end

    %% Backend Layer
    subgraph "Backend Services"
      SYS["Orchestrator<br/>(multimodal_rag/system.py)"]:::processing

      subgraph "Ingestion Pipeline"
        DOC["Document Processor<br/>(PyPDF2, python-docx)"]:::processing
        IMG["Image Processor<br/>(BLIP + OCR)"]:::processing
        AUDIO["Audio Processor<br/>(Whisper)"]:::processing
      end

      subgraph "Query Pipeline"
        EMU["Embedding Service<br/>(Nomic Embed / CLIP)"]:::modeling
        RCV["Retrieval & Context Builder"]:::processing
        LLM["LLM Inference<br/>(Llama3.1 via Ollama)"]:::modeling
      end
    end

    %% Storage Layer
    subgraph "Data Layer"
      DB["SQLite Metadata DB"]:::storage
      DEC{"Choose ChromaDB or FAISS"}:::decision
      CH["ChromaDB Store"]:::storage
      FAISS["FAISS Store"]:::storage
    end

    %% Flows
    Start --> UI
    UI -->|“upload file or query”| SYS

    %% Ingestion flows
    SYS -->|“ingest request”| DOC
    SYS -->|“ingest request”| IMG
    SYS -->|“ingest request”| AUDIO

    DOC -->|“chunks”| EMU
    IMG -->|“chunks”| EMU
    AUDIO -->|“chunks”| EMU

    EMU -->|“embeddings”| DEC
    DEC -->|“Chroma path”| CH
    DEC -->|“FAISS path”| FAISS

    DOC -->|“metadata”| DB
    IMG -->|“metadata”| DB
    AUDIO -->|“metadata”| DB

    %% Query flows
    SYS -->|“query text”| EMU
    EMU -->|“vector”| DEC
    DEC -->|“Chroma search”| CH
    DEC -->|“FAISS search”| FAISS

    CH -->|“retrieved chunks”| RCV
    FAISS -->|“retrieved chunks”| RCV

    RCV -->|“context”| LLM
    LLM -->|“response”| UI

    UI --> End

    %% Click Events
    click UI "https://github.com/itanishqshelar/smartrag/blob/main/chatbot_app.py"
    click SYS "https://github.com/itanishqshelar/smartrag/blob/main/multimodal_rag/system.py"
    click DOC "https://github.com/itanishqshelar/smartrag/blob/main/multimodal_rag/processors/document_processor.py"
    click IMG "https://github.com/itanishqshelar/smartrag/blob/main/multimodal_rag/processors/multimodal_image_processor.py"
    click AUDIO "https://github.com/itanishqshelar/smartrag/blob/main/multimodal_rag/processors/audio_processor.py"
    click CH "https://github.com/itanishqshelar/smartrag/blob/main/multimodal_rag/vector_stores/chroma_store.py"
    click FAISS "https://github.com/itanishqshelar/smartrag/blob/main/multimodal_rag/vector_stores/faiss_store.py"

    %% Styles
    classDef frontend fill:#ABD4FF,stroke:#0366d6;
    classDef processing fill:#CFFFB3,stroke:#228B22;
    classDef storage fill:#FFD8A8,stroke:#FF8C00;
    classDef modeling fill:#E0BBE4,stroke:#800080;
    classDef decision fill:#FFEE93,stroke:#DAA520;
    classDef terminator fill:#FFFFFF,stroke:#000000,stroke-width:2px;
```
