Dan Martín Mecikovsky

A model answers. Someone still has to ask whether the answer is any good.

Philosophy professor (University of Buenos Aires). My question has always been what makes us human and sets us apart from the rest of existence. Now I ask it of the limits and challenges AI puts in front of us — and I build with these tools while I do.

meci.dan@gmail.com Buenos Aires · Remote
Selected work
TEACHING TOOLS · CLASSROOM-TESTED

Interactive lessons

Philosophy you scroll through; exercises that correct themselves.

A set of self-contained HTML/JS tools I build for my own classes: a Descartes scrollytelling lesson, a feudalism-to-modernity timeline, and a guided Plato cave sequence. They are used as classroom material: students scroll, answer, retry, and get immediate feedback without accounts or student data.

HTML/JS Prompt-built Ed-tech Self-correcting
Open the live demo
CREATIVE WRITING · CLASSROOM APP

The Imaginarium Machine

A classroom machine for starting stories and leaving the writing to students.

A local-first creative writing app for A2 English learners. It organizes prompt cards, characters, constraints, and play modes; students use them to write, draw, speak, or act their own stories.

Classroom app Prompt design Creative writing
Open the live demo
AI AGENT · GOVERNMENT DATA

EHOBA

A monthly 53-hotel survey turned into a rule-bound processing agent.

EHOBA processes Buenos Aires's monthly hotel occupancy survey for IDECBA, validates submissions against fixed consistency rules, and follows a documented eight-step correction order. I built it end to end with Claude Code, including the desktop UI, processing pipeline, and design system.

Claude Code Agent design Data processing Design system
Animated scrubbed EHOBA walkthrough with synthetic hotel rows and fictional identifiers Watch the walkthrough
How I work

I've worked with language models since 2022 — prompting them, testing their reasoning, watching where they fail.

I build with them on one condition: they extend human judgment, they don't stand in for it. Bias, gaps, and confident errors are the interesting part — the places where a philosopher's training earns its keep. That criterion is the whole point.