AI-powered research discovery platform that matches your research interests with relevant scientific papers using intelligent semantic ranking.

---

📋 Prerequisites

Tool	Required For	Installation
Docker	All setups	docker.com
Python	3.11.14 — Development setup only	python.org
Node.js	22.12.0+ — Development setup only	nodejs.org

You also need an OpenAI API key for semantic ranking and paper analysis features. Get one at platform.openai.com.

---

🚀 Quick Start (Docker)

Run the entire stack — database, backend, frontend, and test data seeding — in Docker with a few commands.

1. Create your environment file:

   cp .env.example .env

2. Add your OpenAI API key by editing the .env file:

   OPENAI_API_KEY=sk-your-key-here
   

3. Start all services:

   docker compose --profile full up -d

   This will automatically:

   • Start the PostgreSQL database (with pgvector)
   • Launch the backend, which creates the database schema on startup
   • Seed the database with test papers and a test user (test / My Test Project)
   • Serve the frontend via nginx

Once running, open http://localhost in your browser.

Service	URL
Frontend	http://localhost
Backend API	http://localhost:8000
API Docs	http://localhost:8000/docs

To stop all services: docker compose --profile full down

---

🛠️ Development Setup

For active development, run only the database in Docker and start the frontend and backend manually. This gives you hot-reload and direct access to the code.

🖥️ Frontend

1. Navigate to the frontend directory:

   cd frontend

2. Install dependencies:

   npm install

3. Start the development server:

   npm run dev

   The frontend will be available at http://localhost:5173 (or the port shown in the console).

⚙️ Backend

1. From the project root, start the database:

   docker compose up -d

2. Navigate to the backend directory:

   cd backend

3. Set up a Python environment (optional but recommended)

   Choose either a standard venv or an Anaconda environment.

   Option A — venv (Python 3.11.14)

   Ensure Python 3.11.14 is available on your system (python --version or python3 --version).

   Linux / macOS:

   python3.11 -m venv inquiro-env
   source inquiro-env/bin/activate

   Windows (PowerShell):

   py -3.11 -m venv inquiro-env
   inquiro-env\Scripts\activate

   Option B — Anaconda

   conda create -n inquiro-env python=3.11.14
   conda activate inquiro-env

   This ensures all dependencies install cleanly — especially libraries like torch, transformers, and scientific packages.

4. Install dependencies:

   pip install -r requirements.txt
   pip install -r requirements-dev.txt

5. Create a local environment file:

   Copy the example file and rename it to dev.env:

   Linux / macOS:

   cp .env.example dev.env

   Windows (PowerShell):

   copy .env.example dev.env

   Then open dev.env and add your OpenAI API key. Adjust other values if needed (e.g., database port, credentials).

6. Install Git hooks:

   pre-commit install

   After this, formatters and linters will run automatically on every commit.

7. Start the FastAPI server:

   uvicorn app.main:app --reload

   The API will be available at http://localhost:8000 API docs: http://localhost:8000/docs

---

🧪 API Testing with Bruno

The project includes a Bruno collection under /bruno for testing and exploring the backend API. Bruno is a lightweight, file-based API client that stores requests in plain text, making it well-suited for collaborative development and version control.

1. Install Bruno — Download from usebruno.com/downloads.

2. Open the collection — Launch Bruno, click "Open Collection", and select the /bruno/Inquiro Bruno folder from the project root.

3. Select an environment — The /bruno/Inquiro Bruno/environments directory contains predefined environment files. For local development, select the Development environment.

---

🤖 AI Acknowledgement

AI-assisted tools, including ChatGPT, Claude (Code), and Cursor, were used during the development of this project for architectural planning, code generation, and debugging support. All AI-generated output was reviewed and adapted by the development team.