—
Welcome — README
FileEditFontSizeStyleSimpleText

Felipe Guarin

ARTIST & AI LEADER · AI LEAD AT PLATZI · ATLANTA


I'm an artist and AI leader based in Atlanta.

My installation Apperception turns a visitor's brain activity into a living landscape, in real time. It was shown at LOOP Atlanta during SOUND + VISION, the Atlanta Film Festival's 50th anniversary.

By day I'm AI Lead at Platzi, building generative AI products that change how people learn. Before that I led AI at Adtechnacity, where our products drove an 89% increase in RPM and more than $10M in new revenue.

Along the way: 500+ AI projects shipped, an esports platform built in Japan, an award-winning creative agency, and an MIT Hackathon win.

Want to learn AI, build with it, or bring an installation to your space? Let's talk.

— double-click any icon on the desktop to open a program —

Apperception (2026) — Exhibition
ExhibitionLOOP Atlanta · Sept 12, 2026
1 / 29Video from the night, LOOP Atlanta, Sept 12 2026

Apperception (2026)

An interactive brain-to-image installation. Live EEG, generative AI and a real-time WebGL renderer.

ARTIST
Felipe Guarin (concept, software, AI system, live operation)
ART DIRECTION
Catalina Lotero
EXHIBITED
LOOP, Atlanta (665 Marietta St NW), as part of SOUND + VISION, the Atlanta Film Festival's 50th-anniversary event, Saturday, September 12, 2026, 7–11 PM
ON THE NIGHT
More than 40 participants, one at a time, and over 300 images generated live from their brain activity.

STATEMENT

Apperception is the moment perception becomes aware of itself, when you are not only seeing an image but realise you are the one perceiving it.

One visitor at a time sits in front of a large projected landscape wearing a Muse S Athena EEG headband. Their shifting attention, calm, arousal and drift become a live influence on a luminous world: light, bloom, density, fracture and atmosphere. The image is not a brain scan and it does not claim to read thoughts. It is a perceptual terrain, an external image of the act of perceiving, and everyone in the room watches one person's inner state become a shared public environment: body, signal, image, perception, body.

The visual world, developed with Catalina Lotero, is a hyper-real synthetic-organic garden: reflective dark terrain, prismatic light, macro-botanical forms, mineral surfaces and cinematic depth.

HOW IT WORKS

  1. A Muse S Athena headband streams raw EEG over a custom Bluetooth bridge built for the Athena.
  2. An affect pipeline reduces each 45-second "chapter" of clean signal to one of four held brain states: tense (cold, sharp, electric), calm (warm, still, golden), drifting (floating, misty, deep) and open (blooming, branching, multiplying).
  3. Each state draws a visual "gesture" from a 74-gesture vocabulary, so a long-held state keeps evolving instead of looping.
  4. A generative image model (FLUX) edits a founding garden frame: the composition stays fixed and one visible change happens per chapter, so tension can crack the crystal and later calm can heal the same cracks with gold light.
  5. A continuity checker rejects frames that lose the garden's identity, and a WebGL renderer dissolves each new frame in over about 25 seconds, with a slow light tide shaped by the EEG between frames.
  6. An operator console shows headset contact, battery, calibration, the brain trace and spend, so visitors turn over without restarting the piece.
Open: