wizkkk

Applied statistical methods (e.g., regression, clustering) to support client modeling; cleaned and modeled 3,000+ questionnaire responses, achieving a 95% report delivery rate and improving analysis efficiency. Completed full-cycle analysis from data preprocessing and feature engineering to visualization and modeling; independently delivered client reports. Experience in user behavior modeling and statistical analysis is transferable to Web3 user analytics (e.g., Dune dashboards, Swap pattern analysis).

Built multi-table SQL queries to process store and product data, enabling SKU-level inventory tracking across multiple business layers. Designed FineBI dashboards to support multi-dimensional analysis; optimized supply chain links and reduced inventory audit time by ~40%. Independently managed data cleaning, ETL, and indicator modeling; achieved end-to-end delivery from business logic to BI reporting. Translated experience into Web3 user behavior analytics (e.g., retention, on-chain asset activity, transaction pattern tracking) and reusable Dune dashboard strategies.

Developed a Python-based job market analysis system; crawled ~100K job postings from 20 cities, built ETL pipelines, and delivered multi-dimensional visualization dashboards. Built a custom Discuz forum system with automated registration and login modules; supported Selenium-based testing to ensure stable deployment. Achieved top project rating (98/100); the system enables scalable structured data extraction and user behavior modeling, applicable to Web3 on-chain path attribution and interaction analysis.