Michael Zahorec

Philosopher & Computer Scientist. AI Ethics, Explainability, and Evaluation. Philosophy of Science.

Here are some brief descriptions and links. If you are working on a similar probject and want to collaborate, please feel free to reach out to me for more information!

Check out our tool for searching the published works of G.E.M. Anscombe: https://anscombesearch.collegiuminstitute.org/. Built using Elasticsearch. Work funded by the Collegium Institute.

Tetens NEH Project

I am leading the efforts to digitize German versions of the philosophical works, broadly construed, of Johann Nicolaus Tetens (1736-1807). This work follows Text Encoding Initiative (TEI) guidelines. See here for Python XML(TEI)-generator: https://github.com/mzahorec/tetens. This work is in coordination with a new print translation of these philosophical works. For more details: https://tetensproject.create.fsu.edu/.

Corpus Analysis for Experimental Philosophy

In this paper, we developed and deployed new methods in experimental philosophy, which involved analyzing and aggregating linguistic judgments across a wide range of language data, drawn from the British National Corpus. See the paper for our approach, motivation, and potential extensions of this work.

Language Model Evaluation at Humana

During Summer 2024, I worked full-time at Humana with the Responsible AI Team. We leveraged existing frameworks and collections of methods (e.g., HELM and TrustLLM) to improve LLM evaluation processes at Humana in an empirically-informed, principled manner. Primarily concerned with Summarization and Question-Answer tasks.

Data Science at MathWorks

For three summers (2020-2022), I worked full-time at MathWorks with the Data Science Courses Team. I helped develop scripts, projects, assessments, and more. See, e.g., Practical Data Science with MATLAB on Coursera.

Formatting and Index Generator – Critique of Practical Reason

For Courtney Fugate’s new translation of Kant’s Second Critique. In addition to formatting the book in InDesign, I wrote Python code to parse the raw text and generate an index. It implements various replacement rules and rules for term-searching. Feel free to peruse: https://github.com/mzahorec/indexgen.

This Website

If you would like to make a website like this one, feel free to peruse my modifications (https://github.com/mzahorec/mzahorec.github.io) of this Jekyll theme.