Project Overview

This project simplifies data analysis by removing the need to write SQL manually. Users can enter questions in plain English, and the application automatically:

• Converts natural language to SQL using Gemini API
• Executes queries on Snowflake database
• Displays results in tabular and chart format
• Allows users to download results as CSV

This makes data analysis accessible to non-technical users.

---

Technologies Used

• Python
• Streamlit (Web interface)
• Snowflake (Cloud database)
• Gemini API (AI SQL generation)
• Plotly (Data visualization)
• Pandas (Data handling)
• Snowflake Python Connector

---

Features

• Natural language to SQL conversion
• Snowflake database integration
• Query execution and result display
• Automatic chart generation
• CSV download option
• Query history tracking

---

Project Structure

myproject/
│
├── app/
│   ├── main.py                 # Streamlit web app
│   ├── ai_sql_generator.py    # Converts question to SQL using Gemini
│   ├── query_executor.py      # Executes SQL in Snowflake
│   ├── snowflake_client.py    # Snowflake connection
│   ├── visualization.py       # Creates charts
│   ├── history_manager.py     # Saves query history
│   ├── config.py              # Loads environment variables
│
├── data/
│   └── sample_data.py
│
├── .env                       # API keys and credentials
├── requirements.txt
├── run.py
└── README.md

---

System Workflow

User Input (Natural Language)
          │
          ▼
Gemini API generates SQL
          │
          ▼
Snowflake executes SQL
          │
          ▼
Results returned to app
          │
          ▼
Display table + visualization + CSV download

---

Database Schema Example

Table: SALES

Columns:

• order_id
• customer_name
• product
• category
• region
• quantity
• price
• order_date

---

Installation and Setup

1. Clone the repository

git clone <your-repo-url>
cd myproject

2. Create virtual environment

python -m venv venv
venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Configure environment variables

Create .env file:

GEMINI_API_KEY=your_api_key

SNOWFLAKE_USER=your_user
SNOWFLAKE_PASSWORD=your_password
SNOWFLAKE_ACCOUNT=your_account
SNOWFLAKE_WAREHOUSE=your_warehouse
SNOWFLAKE_ROLE=your_role

---

5. Run the application

streamlit run app/main.py

---

Example Usage

User Input:

Show total sales by region

Generated SQL:

SELECT region, SUM(price * quantity) AS total_sales
FROM sales
GROUP BY region;

---

Key Components Explanation

ai_sql_generator.py

Uses Gemini API to convert natural language to SQL.

query_executor.py

Executes SQL query on Snowflake and returns results.

snowflake_client.py

Creates connection to Snowflake database.

visualization.py

Generates charts using Plotly.

main.py

Handles user interface using Streamlit.

---

Problems Solved

Problem: Non-technical users cannot write SQL queries.

Solution: AI automatically generates SQL from natural language.

Problem: Manual data analysis is slow.

Solution: Automated query execution and visualization.

---

Future Improvements

• Support multiple databases
• Add authentication system
• Improve SQL accuracy with schema awareness
• Add dashboard saving feature