Featured interview
“There is a lot of pessimism within the very community that builds AI”
A conversation with Martín De Ambrosio about AI safety, the future of work and the uncertainty of building intelligent machines.
Brands and institutions I’ve worked with
Press & conversations
Interviews, reviews and conversations about artificial intelligence, neuroscience and the ideas behind my work.
A conversation with Martín De Ambrosio about AI safety, the future of work and the uncertainty of building intelligent machines.
A long conversation with Marcelo Longobardi about agent swarms, the paperclip thought experiment, AI’s ethical limits, techno-autocrats and the future of work.
Consciousness, emotions and the place of humans in a world shared with AI.
An interview with Paulino Rodrígues about the risks of AI and how to regulate it.
An interview with Nancy Dahab Martínez about AI’s impact on work, science and health, and what is left for us if we stop being the most intelligent species.
An interview on Spain’s La rosa de los vientos about the debates artificial intelligence has opened among experts and scientists.
Marcelo Longobardi introduces our interview on his show, a conversation about what artificial intelligence is and isn’t.
A conversation with Raúl Cohe about consciousness, emotions and whether an AI can really feel anything.
My appearance on El pase de la tarde. The video opens directly at the interview.
A conversation with Julián Otal Landi and Sofía Aranda González about AI in history education.
A radio conversation about the book and the possible futures of AI.
Fabrizio Sanguinetti reviews the book through the lens of biology, intelligence and consciousness.
The book in the Ideas section of LA NACION.
An interview about the book and how to think about artificial intelligence.
IFIBYNE shares the recognition awarded to the neuroscience podcast A Todo Neuro.
How fish combine sights and sounds to escape danger: the research with Violeta Medan.
Two angles on the same question , how brains and AI models represent ideas. PhD work on the zebrafish brain, postdoc work on what happens inside language models.
Do language models help some people gain power more than others? Across 24 US and Chinese models, whether they refuse depends on who gains, who loses, and who is asking.
Can multimodal LLMs explain their own image classifications? Asking them for formal, concept-based explanations makes their predictions less accurate (93.8%→90.1%).
Asks whether language models can honestly report on their own inner states. Combines AI self-reports with independent measurements to show that, when properly elicited, models give a real readout of how they’re “feeling”.
Asks whether language models build mental maps of space from pure text. Shows they do , and that the same internal map a model uses to navigate one task transfers to other spatial-reasoning tasks.
Whole-brain imaging in zebrafish to watch sight and hearing combine in real time when a threat appears. Custom hardware and analysis software link the brain activity to the actual escape behavior.
How fish combine sight and sound to escape danger faster, especially when each individual signal is weak. A simple model of where the signals merge in the brain explains the behavior.
Society for Neuroscience (SfN)
For the podcast A Todo Neuro for the Argentinian Neuroscience Society.
Ministry of Science · Argentina
For environmental science shorts at Xplora.
Travel Grant
3-month research stay in Paris with Germán Sumbre , IBENS, École Normale Supérieure.
Fulbright Commission
Scholarship for studies at the University of Alabama, USA.
City of Buenos Aires
City of Buenos Aires academic-excellence award.
Code I’ve shipped, open and used by other people.
An open-source toolkit I built to coordinate teams of AI agents , chatbots, decision systems, agents that can use tools. It powers the bot on this page, the CasanChef WhatsApp recipe assistant, and several products my company ships to clients.
Read the docs →
Open-source library that detects what a language model is “thinking about”, and lets researchers gently steer its behavior.
AI products and brand experiences built end-to-end , from the first prototype down to the version that actually reaches the users.
An AI cooking assistant on WhatsApp. Tell it what’s in your fridge and it suggests dishes you can actually make. Built for CasanCrem and used by thousands of people.
A beauty-brand activation that lets users try on different hairstyles in real time, in front of a camera. The first AI Beauty-Tech experience in Argentina, made for Sedal.
A daily dashboard that reads the news and turns Argentine politics and the economy into clear charts. Generated automatically every morning.
Produce a whole video with AI , from script and characters to generated images, video clips and the final edited cut, exported as a finished piece.
A music player where every song, album cover and artist is AI-generated , a glimpse of where music distribution might go in a few years.
A personal assistant on Telegram. Manages your calendar, email, music, food log, and audio notes , all by chatting with it like a person.
Understand any phrase in another language, word by word , translation, pronunciation (IPA), conjugations and cultural context. 28 pairs, by voice or photo.
A daily news bot. Reads a curated feed of voices on X every morning, summarizes the day, and sends the report to your Telegram.
Animated music videos, scripts, and broadcast pieces , produced in-house or as creative direction for partner studios.
A university diploma I created, coordinate and teach at UNLaM: the whole audiovisual pipeline with generative AI, from script to image, video, sound and publishing.
Animated music videos for kids , the Argentine national anthem and other classics, brought to life with AI-assisted animation.
An ongoing series of animated music videos for kids. Original songs, full production, AI-assisted animation.
An animated children’s series celebrating the traditions of Entre Ríos, produced end-to-end with El Perro en la Luna and the provincial government. Visit the official site.
A five-episode spin-off of Los Viajes de Gurisa for the Government of Entre Ríos, with a classroom guide for each episode. I directed the AI-assisted animation with El Perro en la Luna. Visit the official site.
The AI-video agency where I produce work for brands. From the creative idea to the final cut, with AI tools at every step.
Custom dashboards, finance tools, immersive sites, and small businesses online. Designed, built, and maintained end-to-end.
A philosophy audiobook library: the great works of human thought, narrated to listen anywhere. It remembers the exact spot where you left off.
A personal dashboard for tracking my own life , sleep, mood, focus, exercise. The data lives in a simple spreadsheet I can edit anywhere.
A clean, clinical-feel dashboard for personal health metrics. Weight, blood pressure, resting heart rate, sleep, glucose , all in one place.
A site I designed and built for a small home bakery. Catalog, gallery, and an easy way for customers to order custom cakes.
A piece of the Paraná Delta you can explore in your browser, scanned in 3D and rebuilt as an immersive walk-through. A love letter to a real ecosystem.
Music, podcasts, science outreach. Where the technical work meets the public conversation.
A science communication channel I co-founded. Weekly videos for C5N, animated shorts for Canal Encuentro. Won the “Ciencia por Contar” government award in 2023.
An award-winning neuroscience podcast for the Argentine Society of Neuroscience. Won the SfN Next Generation Award in 2024.
Original songs across genres , writing, recording, playing several instruments. Personal and collaborative projects.
I created, coordinate and teach this university diploma: the whole audiovisual pipeline with generative AI.
Lead Teaching Assistant in Computational Neuroscience and Deep Learning, in both English and Spanish.
Advisor on the new primary curriculum and AI integration; educator trainer in Natural Sciences across 16 curricular areas.
For corporations, schools, ministries, and government teams. Hands-on, with custom curricula tailored per audience.
Head of Research at Buenos Aires AI Safety Hub.