Built an end-to-end pipeline from scratch: a Selenium scraper that gets past Cloudflare and infinite-scroll pagination to collect 4,200+ titles, a reproducible 5-step cleaning pipeline (down to 1,648 records), and three analyses — a Random Forest rating classifier (ROC-AUC 0.87) with leakage-aware feature engineering and stratified cross-validation, K-means clustering, and a Mann-Whitney U hypothesis test.

Turned dense PDFs into a conversational study tool: a RAG pipeline (Next.js 13) with recursive text splitting, text-embedding-ada-002 embeddings, and PineconeDB semantic retrieval, plus chat-history condensation that rewrites queries to preserve context across multi-turn sessions.

Mapped 25,000 Spanish pharmacies into an interactive tool for client segmentation: a Python scraper captured location, contact, and operational metadata from government sources with dynamic content handling and validation, feeding a MapBox GL JS interface with clustering and filtering over the geospatial distribution.