saurabhanand/AI-Powered_SQL_Business_Intelligence_Assistant
0
🤖 AI-Powered SQL Business Intelligence Assistant
Ask business questions in plain English → AI generates SQL → results + charts + insights
     
🔴 Live Demo
📌 What This Project Does
A user types: "Which region has the highest profit margin?"
The app:
- Sends the question to Gemini AI API
- AI generates a valid SQL query
- Query runs on a SQLite database (9,994 orders)
- Results displayed as table + auto chart
- AI generates a plain English business insight
🗂️ Project Structure
AI-SQL-Business-Intelligence-Assistant/
│
├── app.py # Main Streamlit application (4 pages)
├── superstore.csv # Superstore sales dataset (9,994 rows)
├── superstore.db # SQLite database (auto-created from CSV)
├── requirements.txt # Python dependencies
└── README.md📁 Dataset
Columns
🌐 App Pages
🛠️ Tech Stack
🚀 Run Locally
# 1. Clone the repo
git clone https://github.com/SaurabhAnand56/AI-SQL-Business-Intelligence-Assistant.git
cd AI-SQL-Business-Intelligence-Assistant
# 2. Install dependencies
pip install -r requirements.txt
# 3. Run the app
streamlit run app.pyGet a free Anthropic API key at console.anthropic.com and enter it in the app.
🔍 SQL Examples Used
-- Year-over-year growth
SELECT strftime('%Y', order_date) as year,
ROUND(SUM(sales), 2) as total_sales,
COUNT(DISTINCT order_id) as orders
FROM sales
GROUP BY year ORDER BY year;
-- Profit margin by category
SELECT category,
ROUND(SUM(profit)/SUM(sales)*100, 2) as margin_pct
FROM sales GROUP BY category;
-- Window function: running total
SELECT order_date,
ROUND(SUM(SUM(sales)) OVER (ORDER BY order_date), 2) as running_total
FROM sales GROUP BY order_date ORDER BY order_date;📬 Connect
 