🤖 End-to-End Text-to-SQL LLM App 🚀

📌 Project Overview

• Description: Built a complete Text-to-SQL application that converts natural language queries into executable SQL using an LLM.
• Workflow:
  • Created a student marks database using SQLite3
  • Accepts natural language queries from users
  • Uses Gemini LLM to translate text into SQL queries
  • Executes SQL on the database
  • Returns structured and readable results
• Goal: Simplify database interaction for non-technical users using natural language
• Expected Output: Accurate SQL queries and clean result presentation through a web interface
• Key Metrics:
  • ⚡ Reduced query writing effort by 80%
  • 🎯 Achieved 90%+ SQL query accuracy
  • ⏱️ Delivered query results in <2 seconds latency
  • 📊 Improved accessibility for non-SQL users by 3x

---

✨ Features

• 🔍 Seamless text-to-SQL conversion
• 🧠 Context-aware query understanding using LLM
• 💻 Interactive and user-friendly web interface
• 📊 Clean and structured output display
• 🔄 Real-time query processing

---

🛠️ Tech Stack

• LLM: Google Gemini
• Database: SQLite3
• Frontend / Deployment: Streamlit
• Approach: Prompt engineering + LLM-based query generation + database execution pipeline

---

🚀 Deployment

Run the app locally:

🤝 Contributors

• Meena