# msvivekranjan/studymitra architecture

> Architecture diagram of the GitHub repository msvivekranjan/studymitra, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-19). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/msvivekranjan/studymitra
- Repository: https://github.com/msvivekranjan/studymitra

## Overview

StudyMitra is a personal academic RAG study assistant. Users add documents and metadata through the library, which stores and indexes them; in chat, they select a response mode, ask a question, and receive a streamed, cited answer grounded in retrieved sources. The graph covers the documented ingestion, hybrid retrieval, generation, and library capabilities. Some pipeline internals and API wiring are unsampled, so their relationships are omitted or marked unverified.

Main entry points: [`backend/app/ingestion/pipeline.py`](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/pipeline.py) (Ingestion pipeline), [`backend/app/retrieval/retriever.py`](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/retriever.py) (Retrieval orchestrator), [`frontend/src/App.jsx`](https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/App.jsx) (React workspace), [`backend/app/generation/generator.py`](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/generator.py) (Grounded answer generator)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_client["Study workspace"]
  node_workspace["React workspace<br/>[App.jsx]"]
  node_library_ui["Document library<br/>[Library.jsx]"]
  node_upload_ui["Upload panel<br/>[UploadPanel.jsx]"]
  node_modes["Study modes<br/>[studyModes.js]"]
  node_api_client["API and SSE client<br/>[api.js]"]
end

subgraph group_api["Backend services"]
  node_document_api["Document API<br/>[documents.py]"]
  node_chat_api["Chat API<br/>[chat.py]"]
  node_document_service["Document service"]
  node_chat_service["Chat service<br/>[chat_service.py]"]
  node_metadata_db[("Metadata database<br/>[database.py]")]
end

subgraph group_ingestion["Document ingestion"]
  node_ingestion_pipeline["Ingestion pipeline<br/>[pipeline.py]"]
  node_file_loader["Document loaders<br/>[loader.py]"]
  node_chunker["Structure-aware chunker<br/>[chunker.py]"]
  node_metadata["Metadata enrichment<br/>[metadata.py]"]
end

subgraph group_retrieval["Search and grounding"]
  node_embedder["Local embeddings<br/>[embedder.py]"]
  node_vector_store[("Qdrant vector store<br/>[vector_store.py]")]
  node_keyword_search["BM25 keyword search<br/>[keyword.py]"]
  node_hybrid_merge["RRF fusion<br/>[hybrid.py]"]
  node_reranker["Cross-encoder reranker<br/>[reranker.py]"]
  node_retriever["Retrieval orchestrator<br/>[retriever.py]"]
end

subgraph group_generation["Answer generation"]
  node_generator["Grounded answer generator<br/>[generator.py]"]
  node_prompts["Mode prompt templates<br/>[prompts.py]"]
  node_llm["LLM provider client<br/>[llm.py]"]
end

node_student(("Student"))
node_llm_provider{{"Chosen LLM provider"}}

node_student -->|"asks questions"| node_workspace
node_student -->|"manages resources"| node_library_ui
node_workspace -->|"sends requests"| node_api_client
node_workspace -->|"renders"| node_library_ui
node_workspace -->|"offers modes"| node_modes
node_library_ui -->|"renders"| node_upload_ui
node_api_client -->|"calls document routes"| node_document_api
node_api_client -->|"streams chat"| node_chat_api
node_document_api -->|"delegates uploads"| node_document_service
node_document_api -->|"reads document"| node_metadata_db
node_document_service -->|"writes metadata"| node_metadata_db
node_document_service -->|"ingests document"| node_ingestion_pipeline
node_ingestion_pipeline -.->|"processes files"| node_file_loader
node_ingestion_pipeline -.->|"creates chunks"| node_chunker
node_ingestion_pipeline -.->|"enriches chunks"| node_metadata
node_ingestion_pipeline -.->|"embeds chunks"| node_embedder
node_ingestion_pipeline -.->|"stores vectors"| node_vector_store
node_chat_api -.->|"dispatches queries"| node_chat_service
node_chat_service -->|"generates answers"| node_generator
node_generator -.->|"gets context"| node_retriever
node_retriever -->|"embeds query"| node_embedder
node_retriever -->|"searches vectors"| node_vector_store
node_retriever -->|"searches terms"| node_keyword_search
node_retriever -->|"merges results"| node_hybrid_merge
node_retriever -->|"reranks results"| node_reranker
node_generator -.->|"formats prompts"| node_prompts
node_generator -.->|"requests completion"| node_llm
node_llm -->|"sends prompts"| node_llm_provider

click node_workspace "https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/App.jsx"
click node_library_ui "https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/features/Library.jsx"
click node_upload_ui "https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/features/UploadPanel.jsx"
click node_modes "https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/lib/studyModes.js"
click node_api_client "https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/lib/api.js"
click node_document_api "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/api/documents.py"
click node_chat_api "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/api/chat.py"
click node_document_service "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/services/document_service.py"
click node_chat_service "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/services/chat_service.py"
click node_metadata_db "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/db/database.py"
click node_ingestion_pipeline "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/pipeline.py"
click node_file_loader "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/loader.py"
click node_chunker "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/chunker.py"
click node_metadata "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/metadata.py"
click node_embedder "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/embedder.py"
click node_vector_store "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/vector_store.py"
click node_keyword_search "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/keyword.py"
click node_hybrid_merge "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/hybrid.py"
click node_reranker "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/reranker.py"
click node_retriever "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/retriever.py"
click node_generator "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/generator.py"
click node_prompts "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/prompts.py"
click node_llm "https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/llm.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_workspace,node_library_ui,node_upload_ui,node_modes,node_api_client toneBlue
class node_document_api,node_chat_api,node_document_service,node_chat_service,node_metadata_db toneAmber
class node_ingestion_pipeline,node_file_loader,node_chunker,node_metadata toneMint
class node_embedder,node_vector_store,node_keyword_search,node_hybrid_merge,node_reranker,node_retriever toneRose
class node_generator,node_prompts,node_llm,node_student,node_llm_provider toneIndigo
```

## Components

### Study workspace

- [React workspace](https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/App.jsx): `frontend/src/App.jsx`
- [Document library](https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/features/Library.jsx): `frontend/src/features/Library.jsx`
- [Upload panel](https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/features/UploadPanel.jsx): `frontend/src/features/UploadPanel.jsx`
- [Study modes](https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/lib/studyModes.js): `frontend/src/lib/studyModes.js`
- [API and SSE client](https://github.com/msvivekranjan/studymitra/blob/main/frontend/src/lib/api.js): `frontend/src/lib/api.js`

### Backend services

- [Document API](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/api/documents.py): `backend/app/api/documents.py`
- [Chat API](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/api/chat.py): `backend/app/api/chat.py`
- [Document service](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/services/document_service.py): `backend/app/services/document_service.py`
- [Chat service](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/services/chat_service.py): `backend/app/services/chat_service.py`
- [Metadata database](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/db/database.py): `backend/app/db/database.py`

### Document ingestion

- [Ingestion pipeline](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/pipeline.py): `backend/app/ingestion/pipeline.py`
- [Document loaders](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/loader.py): `backend/app/ingestion/loader.py`
- [Structure-aware chunker](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/chunker.py): `backend/app/ingestion/chunker.py`
- [Metadata enrichment](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/ingestion/metadata.py): `backend/app/ingestion/metadata.py`

### Search and grounding

- [Local embeddings](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/embedder.py): `backend/app/retrieval/embedder.py`
- [Qdrant vector store](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/vector_store.py): `backend/app/retrieval/vector_store.py`
- [BM25 keyword search](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/keyword.py): `backend/app/retrieval/keyword.py`
- [RRF fusion](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/hybrid.py): `backend/app/retrieval/hybrid.py`
- [Cross-encoder reranker](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/reranker.py): `backend/app/retrieval/reranker.py`
- [Retrieval orchestrator](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/retrieval/retriever.py): `backend/app/retrieval/retriever.py`

### Answer generation

- [Grounded answer generator](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/generator.py): `backend/app/generation/generator.py`
- [Mode prompt templates](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/prompts.py): `backend/app/generation/prompts.py`
- [LLM provider client](https://github.com/msvivekranjan/studymitra/blob/main/backend/app/generation/llm.py): `backend/app/generation/llm.py`

### Other components

- **Student**
- **Chosen LLM provider**

## Connections

- Student → React workspace: asks questions (evidence: `README.md`)
- Student → Document library: manages resources (evidence: `README.md`)
- React workspace → API and SSE client: sends requests (evidence: `frontend/src/App.jsx`)
- React workspace → Document library: renders (evidence: `frontend/src/App.jsx`)
- React workspace → Study modes: offers modes (evidence: `frontend/src/App.jsx`)
- Document library → Upload panel: renders (evidence: `frontend/src/features/Library.jsx`)
- API and SSE client → Document API: calls document routes
- API and SSE client → Chat API: streams chat
- Document API → Document service: delegates uploads (evidence: `backend/app/api/documents.py`)
- Document API → Metadata database: reads document (evidence: `backend/app/api/documents.py`)
- Document service → Metadata database: writes metadata (evidence: `backend/app/services/document_service.py`)
- Document service → Ingestion pipeline: ingests document (evidence: `backend/app/services/document_service.py`)
- Ingestion pipeline → Document loaders: processes files
- Ingestion pipeline → Structure-aware chunker: creates chunks
- Ingestion pipeline → Metadata enrichment: enriches chunks
- Ingestion pipeline → Local embeddings: embeds chunks
- Ingestion pipeline → Qdrant vector store: stores vectors
- Chat API → Chat service: dispatches queries
- Chat service → Grounded answer generator: generates answers (evidence: `backend/app/services/chat_service.py`)
- Grounded answer generator → Retrieval orchestrator: gets context (evidence: `README.md`)
- Retrieval orchestrator → Local embeddings: embeds query (evidence: `backend/app/retrieval/retriever.py`)
- Retrieval orchestrator → Qdrant vector store: searches vectors (evidence: `backend/app/retrieval/retriever.py`)
- Retrieval orchestrator → BM25 keyword search: searches terms (evidence: `backend/app/retrieval/retriever.py`)
- Retrieval orchestrator → RRF fusion: merges results (evidence: `backend/app/retrieval/retriever.py`)
- Retrieval orchestrator → Cross-encoder reranker: reranks results (evidence: `backend/app/retrieval/retriever.py`)
- Grounded answer generator → Mode prompt templates: formats prompts (evidence: `README.md`)
- Grounded answer generator → LLM provider client: requests completion (evidence: `README.md`)
- LLM provider client → Chosen LLM provider: sends prompts (evidence: `backend/app/generation/llm.py`)
