A prototype application which is deployed on Streamlit Community Cloud that transforms natural language into SQL queries, allowing users to explore a company's database using everyday English.
(https://nl2sql-prototype.streamlit.app/)
- 🔐 User registration & login system (PostgreSQL + bcrypt)
- 📋 Consent form signing with auto-generated PDF
- ☁️ PDF storage to AWS S3
- 🧪 Questionnaire feedback module
- 🗃️ Query interface for company data (using SQLite)
- 🖥️ CI/CD support with Streamlit Cloud
-
Register
- Click on "New Register"
- Fill in your first name, last name, username, and password
- Click "Register" to create your account
-
Login
- Enter your username and password
- You'll be redirected to the dashboard
-
Ask Questions
- Type a question like:
- “Show all employees in the marketing department”
- “Total sales by employee last month”
- View the SQL and results directly!
- Type a question like:
-
View Schema
- Expand the schema viewer to understand what tables and columns are available
- Users are shown a consent page when giving feedback
- A PDF of their agreement is generated and uploaded securely to the cloud
- Feedback is collected via a short questionnaire
| Layer | Technology |
|---|---|
| Frontend | Streamlit |
| Backend | Python (fpdf, psycopg2, boto3) |
| Database | PostgreSQL (for users & feedback) + SQLite (for company.db) |
| Storage | AWS S3 |
| Hosting | Streamlit Cloud |
- “List all employees in marketing.”
- “What is the total sales for each department?”
- “Show employees who joined after 2022.”
- “Top 3 sales employees by revenue.”
.
├── assests/ # static files, images etc.
├── backend/ # Auth, registration, database logic
├── database/ # Local SQLite files (e.g., company.db)
├── utils/ # Utility functions (PDF generation, S3 upload, Handle pages etc.)
├── views/ # Streamlit UI pages: login, register, consent, etc.
├── main.py # App entry point
├── requirements.txt # Python dependencies
└── README.md