AutoPrepAI

Developed AutoPrepAI, an AI-powered web application for automated and intelligent data preprocessing, designed to simplify dataset cleaning and preparation for machine learning and data analysis.

The system utilizes six specialized AI agents to handle data standardization, duplicate detection, outlier removal, missing value imputation, feature engineering, and feature selection. It supports automated, manual, and natural language-based preprocessing workflows, allowing users to customize operations and interact with the system using plain English commands.

Implemented an interactive Human-in-the-Loop workflow that enables users to review, approve, or reject suggested preprocessing operations, with detailed explanations and action logs to ensure transparency and explainability.

**Technologies:** Python, FastAPI, React.js, PostgreSQL, Scikit-learn, DSPy, Groq API, SQLAlchemy, Backblaze B2.