🚀 CodeLens AI

AI-Powered Programming & Database Learning Platform

CodeLens AI is an AI-powered learning platform that helps students master Python programming and SQL through interactive analysis, visualization, and intelligent tutoring.

About

CodeLens AI combines deterministic program analysis with AI-powered explanations to make programming concepts easier to understand, debug, and explore.

The platform includes two learning environments:

• Python Lab for code analysis, runtime visualization, complexity metrics, fixes, and AI explanations.
• SQL Lab for schema-aware SQL generation, validation, execution, and tutoring.

Preview

Landing Page

Labs

Python Lab	SQL Lab

Features

🐍 Python Lab

• Static Semantic Analysis
• Runtime Visualization
• Complexity Metrics
• AI Code Explanation
• Suggested Code Fixes
• Interactive Learning

🗄 SQL Lab

• Natural Language → SQL
• Schema-aware SQL Generation
• SQL Validation
• SQL Execution
• AI SQL Tutor
• Dataset Explorer
• Result Visualization

Technology Stack

Layer	Technology
Frontend	Next.js 15, TypeScript, Tailwind CSS, React Query, Monaco Editor, React Flow
Backend	FastAPI, Pydantic, NetworkX, Python 3.11+
SQL Engine	SemanticSQL, FastAPI, SQLAlchemy, sqlglot
AI Runtime	Ollama
AI Models	qwen3:4b, llama3.1:8b, qwen2.5:7b
Database	SQLite

Project Structure

CodeLens-AI/
├── backend/          # Python Lab API, analyzer, runtime, AI explanations
├── sql_backend/      # SQL Lab API, SQL generation, validation, tutoring
├── frontend/         # Next.js application
├── setup.sh          # One-time local setup helper
├── cleanup.sh        # Optional helper for freeing occupied local ports
├── run.sh            # Starts local development services
├── stop.sh           # Stops services started by run.sh
├── status.sh         # Shows local service status
└── README.md

Quick Start

git clone https://github.com/dhruvak99/CodeLens-AI.git
cd CodeLens-AI

chmod +x setup.sh cleanup.sh run.sh stop.sh status.sh

./setup.sh

Start Ollama in a separate terminal:

ollama serve

Run the application:

./cleanup.sh      # Optional, frees occupied CodeLens AI ports
./run.sh

Use cleanup.sh only when ports 3000, 8000, or 8001 are already occupied by stale CodeLens AI processes.

Stop the application:

./stop.sh

Check service status:

./status.sh

Open the application:

http://localhost:3000

Note: For the best experience, open CodeLens AI in an Incognito/Private browsing window. This avoids browser extension interference and cached editor state that can occasionally affect the Monaco editor.

Manual Setup

If you do not want to use the helper scripts, you can manually start each service:

• Backend: create backend/.venv, install backend dependencies, then run FastAPI on port 8000.
• SQL Backend: create sql_backend/.venv, install SQL backend dependencies, then run FastAPI on port 8001.
• Frontend: run npm install inside frontend/, then start Next.js on port 3000.

AI Models

Purpose	Model
Python code explanations	qwen3:4b
SQL generation	llama3.1:8b
SQL tutoring explanations	qwen2.5:7b

Roadmap

Completed

• Python Lab
• SQL Lab
• AI SQL Tutor
• Runtime Visualization
• Schema-aware SQL Generation

Planned

• User Authentication
• Progress Tracking
• Learning Dashboard
• PostgreSQL Support
• Additional Programming Languages
• Docker Support
• Cloud Deployment

Contributing

Contributions are welcome. If you want to improve CodeLens AI, open an issue, propose an enhancement, or submit a pull request with a focused change.

License

This project is licensed under the terms described in the repository LICENSE file.