PageSense
A document-understanding pipeline that detects page structure and extracts text from PDFs and scanned images.
CPU prototype tested
A place to turn ambitious ideas into working software — exploring document AI, application backends, and practical cloud infrastructure.
Real experiments, measured results, and the occasional spectacular dependency conflict.
A document-understanding pipeline that detects page structure and extracts text from PDFs and scanned images.
An examination platform concept for authoring questions, delivering practice tests, and tracking results.
Hands-on Linux, networking, containers, deployment, observability, and secure software operations.