Recent work: a Python scraping pipeline that collects multi-page data into a clean dataset, an exploratory analysis that turns messy tables into clear charts and takeaways, and an image annotation workflow with keypoint and bounding-box quality checks. The through-line is data quality — collecting it, cleaning it, labelling it, and documenting it so the next model or report can trust it. If you need help with scraping, analysis, or computer-vision datasets, send me a message.