About
I'm a Software Developer with over a decade experience in mobile and enterprise systems, from early discovery through production. I approach engineering as a craft, valuing human authorship, collaboration and responsible AI-assisted workflows over black-box automation and proprietary lock-in. Open source, inclusive design, data privacy, digital sovereignty, and sustainable architecture are the principles that guide that practice.
Informed by critical theory and the philosophy of technology, I ground my engineering practice in responsible design, critical thinking and constructive problem-solving: healthy skepticism over convenience, ethics over expediency, and solutions that empower the diverse communities and ecologies that sustain us.
Concept drift is what happens when the complex phenomenon of the world changes after a model is fixed: the statistical relationship between inputs and outputs shifts, assumptions fall out of alignment, and outputs degrade. Model collapse is the terminal form of that failure: when AI systems train on AI-generated data, outputs converge, variance narrows, and the model consumes its own tail.
Whether or not model collapse proves technically inevitable, it serves as a useful metaphor for a society administered by the logic of AI: algorithmic monocultures that attempt to erase the particularities, contradictions, and ambiguities of lived existence. In machine learning, concept drift is a problem to be managed. In the real world, it is what makes us human.
When systems exclude human authorship, they deny contingency, erase difference, and stifle creativity, and with them foreclose the possibility of technology that genuinely serves the many, not just the few. Keeping human authorship legible, accountable, and celebrated is not an empty romantic gesture; it is a technical and ethical necessity.
Technology that augments rather than displaces human agency and creativity is worth creating together.
Get in touch: concept.drift@proton.me
Some Inspiration:
"As first principle … the time saved by automatization must be invested in new capacities for dis-automatization, that is, for the production of negentropy … The time liberated by the end of work must be put at the service of an automated culture, but one capable of producing new value and of reinventing work." — Bernard Stiegler, 2016.