DocAssistant

DocAssistant is a full-stack document intelligence SaaS prototype. Users can upload documents, organize them by workspace, ask questions over document content, and manage AI-assisted document workflows from a dashboard.

Highlights

• FastAPI backend with modular routers for authentication, workspaces, documents, chat, and admin flows
• Next.js dashboard with protected routes for documents, chat, workspace settings, and admin views
• Document ingestion services for PDF/text extraction, chunking, embeddings, and vector search
• PostgreSQL persistence with SQLAlchemy models and seed/init SQL
• Qdrant vector search, Redis, and optional S3-compatible storage through Docker Compose
• JWT authentication, role-aware workspace concepts, and document status tracking
• Streaming-style chat flow designed around retrieval augmented generation

Tech Stack

Area	Tools
Frontend	Next.js, React, TypeScript, CSS modules/global CSS
Backend	FastAPI, Python, SQLAlchemy, Pydantic
AI and search	OpenAI, LangChain, Qdrant
Data and infra	PostgreSQL, Redis, Docker Compose, Nginx
Document processing	PyMuPDF, pdfplumber, pytesseract, Pillow

Project Structure

backend/
  app/
    api/v1/          Auth, document, chat, workspace, and admin routes
    core/            Config, database, and security utilities
    models/          SQLAlchemy entities
    schemas/         Pydantic request/response models
    services/        Document processing, embeddings, storage, LLM, vector store
frontend/
  app/               Next.js app routes and dashboard pages
  lib/               API client, auth, workspace state
infrastructure/
  nginx/             Reverse proxy config
  postgres/          Database initialization
docker-compose.yml   Local multi-service environment

Run Locally

Copy the environment template and fill in local values:

cp .env.example .env

Start the full stack:

docker compose up --build

Useful local URLs:

• Frontend: http://localhost
• API docs: http://localhost:8000/api/docs
• Qdrant: http://localhost:6333

For separate development:

cd backend
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000

cd frontend
npm install
npm run dev

Portfolio Notes

This project is best presented as a document AI platform prototype: dashboard UX, RAG-oriented backend services, workspace-based data modeling, and Dockerized infrastructure. Add screenshots for the dashboard, document upload flow, and chat page before linking it from the portfolio.