Faculty at University of Pennsylvania's Department of Computer & Information Sciences. I love to teach, to mentor and advise students, to think "at scale", to build stuff open source, and to expand the circle of people who identify as "programmers."

• 🔭 I’m currently working on music digital humanities project + CS education, code grading/teaching projects
• 🌱 I’m currently learning TypeScript/React/front-end + machine learning
• 👯 I’m looking to collaborate on open-source projects, especially that reduce the friction to building
• 💬 Ask me about scaling, academic peer review, gamification, centralization/decentralization, capitalism, good software engineering practices, veganism 🐮

• 🎶 imslp: A Python package to query and retrieve scores from the International Music Score Library Project (IMSLP).

• 🎼 incipit: A Python package and command line tool to slice a musical score into bars, staves and systems. Was originally designed to extract the first line of each of Domenico Scarlatti's 555 sonatas to create a searchable catalog with incipit.

You can also visit the GitHub organization of the Domenico Scarlatti Foundation.

• ☢️ React Templates

  • Front-end only: react-ts-starter (React/TypeScript/GitHub Pages)
  • Front-end + database + login: supabase-react-example (React/JavaScript/GitHub Pages/Supabase)

• 🪚 Scraper template: basic-git-scraper-example (Python/GitHub Actions)

• 🕸️ Website templates (Hugo/GitHub Actions/GitHub Pages):

  • https://github.com/jlumbroso/hugo-github-example
  • https://github.com/jlumbroso/hugo-geekdoc-github-example
  • https://github.com/jlumbroso/hugo-timeline-example
  • https://github.com/jlumbroso/hugo-github-bearblog-template
  • https://github.com/jlumbroso/hugo-theme-bootstrap-skeleton
  • https://github.com/jlumbroso/hugo-apero-github-template

• 📧 Google Spreadsheet Mail Merge: An easy-to-use template for no-coders to send mail merges with SendGrid from a list of recipients that is contained in a Google Spreadsheet.

• 🌊 Many data streaming probabilistic algorithms, including those I design and study, use families of hash functions. Hard to find families with good properties (simple, efficient, not too correlated). A affine transform of CRC32 hash, with factors drawn from Mersenne Twister provides a good empirical family. Details are tricky to get right—so I get them right for you!

  • Python implementation at python-random-hash and available on PyPI; Java implementation available at java-random-hash, on Maven Central and GitHub Packages.

• 🙆🏼 Affirmative Sampling (2022) with Conrado Martínez (PDF), is a novel probabilistic sampling algorithm of which the size of the sample grows as a function of the (unknown) number of distinct elements, making it uniquely adaptive to queries that depend on the relative proportion of elements. Reference implemented in Python at affirmative-sampling

• HyperBitBit (2024) with Bob Sedgewick and Svante Janson, is a successor to HyperLogLog that uses half the memory: https://github.com/robert-sedgewick/hyperbitbit