Lab

Workbench

  • Beyond the Metrics: Lessons from NLP Model Evaluation (2024)

    From the archives: my very first foray into model evaluation. Spoiler: even simple-sounding tasks are anything but. Unfortunately, more relevant than ever.

    May 26, 2026

  • Content Capture Bookmarklets

    One-click capture tools for saving content to a personal data lake.

    May 20, 2026

  • Agentic Starter Kit

    A pair of agent skills for personal needs analysis and minimalist solution design.

    May 15, 2026

  • The making of this site

    The technical architecture of this site: Astro, Markdown-first content, and a custom knowledge graph.

    May 15, 2026