LUMEN | Data Intelligence Agent

🚀 View Live Demo →

Bridging the gap between complex databases and actionable business insights.

LUMEN | Data Intelligence Agent is an advanced Text-to-SQL Agent designed to demonstrate the power of Generative AI in Enterprise Data Analytics. It allows non-technical stakeholders to query complex SQL databases using natural language, receive accurate answers, and visualize trends instantly.

---

🚀 Key Features

• Natural Language to SQL: Translates business questions (e.g., "What is the revenue trend for Laptop Pro X?") into optimized SQL queries.
• Autonomous Agent Architecture: Built with LangChain and OpenAI/OpenRouter, the agent understands schema, joins tables, and executes queries securely.
• Interactive Visualizations: Automatically generates Plotly charts for data-heavy responses, enabling instant trend analysis.
• Modern Tech Stack:
  • Frontend: Next.js 14 (App Router), TypeScript, Tailwind-free CSS Modules (Glassmorphism Design).
  • Backend: FastAPI, Uvicorn, SQLite (extensible to PostgreSQL/Snowflake).
  • Tooling: FastMCP (Model Context Protocol) for standardized tool execution.

---

💼 Business Value & Use Case

In the consulting and enterprise world, data accessibility is often a bottleneck. Decision-makers rely on data analysts to write SQL queries, creating delays.

LUMEN | Data Intelligence Agent solves this by:

1. Democratizing Data Access: Empowering managers to get answers instantly without knowing SQL.
2. Reducing Time-to-Insight: From question to visualized answer in seconds, not days.
3. Scalable Architecture: The agentic design allows for easy integration of new data sources and tools.

---

⚠️ Important: Running the Application

The GitHub Pages deployment is a static showcase only. To use the full "Chat with your Data" functionality, you must run the backend locally, as it requires a Python environment to execute SQL queries and generate plots.

Prerequisites

• Python 3.11+
• Node.js 18+
• An API Key (OpenAI or OpenRouter)

1. Clone the Repository

git clone https://github.com/leonleidner/text-to-sql-agent.git
cd text-to-sql-agent

2. Start the Backend (Python)

The backend handles the LLM logic, SQL execution, and plotting.

# Install dependencies
pip install -r requirements.txt

# Start the FastAPI server
# This will run on http://localhost:8001
uvicorn backend.api:app --reload --port 8001

3. Start the Frontend (Next.js)

The frontend provides the chat interface and dashboard.

cd frontend

# Install dependencies
npm install

# Start the development server
# This will run on http://localhost:3000
npm run dev

4. Use the Agent

1. Open http://localhost:3000 in your browser.
2. Scroll down to the Interactive Preview.
3. Click the Settings icon and enter your API Key.
4. Enable "Enable Plotting".
5. Ask a question like:
   • "Show me the monthly revenue for Laptop Pro X over the last 2 years."
   • "Which product has the highest profit margin?"

---

🏗️ Architecture

graph TD
    User[User] -->|Natural Language| Frontend[Next.js Frontend]
    Frontend -->|API Request| Backend[FastAPI Backend]
    Backend -->|LangChain| Agent[SQL Agent]
    Agent -->|Schema Info| DB[(SQLite Database)]
    Agent -->|Generate Plot| PlotTool[Plotly Tool]
    Agent -->|SQL Query| DB
    DB -->|Raw Data| Agent
    Agent -->|Final Answer + JSON| Backend
    Backend -->|Response| Frontend

Loading

---

📬 Contact

Leon Leidner

• GitHub
• LinkedIn

Built as a showcase for advanced Data Engineering & AI Agent capabilities.