Enterprise-grade CSV data quality analyzer powered by Machine Learning. Automatic anomaly detection, statistical profiling, PII scanning, and actionable insights. Secure user authentication, custom data pipelines, and interactive dashboards. Production-ready SaaS application.

This project automates exploratory data analysis (EDA) with DataPulse, enabling users to upload, clean, and visualize datasets effortlessly. It integrates machine learning models like Logistic Regression and XGBoost for insightful analysis via an intuitive Streamlit interface.

Generate valuable insights from customer and transactions data.

A comprehensive end-to-end data warehouse project using MYSQL, covering ETL pipelines, data modeling, and analytics/reporting.

Real-world open data analysis using Excel pivots, visuals, and insights.

A data cleaning project using Microsoft SQL Server to standardize 56,000+ rows of Nashville Housing data. It demonstrates advanced T-SQL techniques (including self-joins, string parsing, and CTEs) to transform messy, unformatted records into a structured dataset ready for analysis.

Анализ эффективности рекламных кампаний и проверка корректности атрибуции пользователей. Проект включает работу с маркетинговыми данными: установки пользователей, рекламные расходы, доходы и каналы привлечения.

This project focuses on analyzing student academic performance data to identify factors that influence exam scores and overall achievement. Using Python and Pandas, the goal is to clean, explore, and analyze the dataset to answer important questions about student performance, study habits, attendance, and other contributing factors.

In this our project we aimed to gather and analyze detailed information on apps in the Google Play Store in order to provide insights on app features and the current state of the Android app market.

Deterministic data-cleaning pipeline for business leads, with a local LLM fallback for messy records. Memory-safe chunking, CSV/DOCX support, and CSV-injection sanitization.

Matelda, an interactive system for multi-table error detection that combines automated error detection with human-in-the-loop refinement.