Join our virtual meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Descubre como puedes aprovechar los modelos de AI para ejecutarlos directamente en tu dispositivo, ofreciendo velocidad, autonomía y privacidad real: los datos nunca abandonan tu teléfono. En esta charla veremos cómo funciona, qué posibilidades abre y por qué marca un antes y un después en la relación entre usuario y tecnología.

Video generation is no longer limited to cloud platforms like Google Veo3 and others!
You can also do it with open-source software—no cost and no restrictions!

In this talk, you’ll discover the node-based interface ComfyUI with models like Wan2.1. I’ll also give you a quick demo of an open-source extension I developed, which has gained huge popularity in the community in just one month (1.2k stars on GitHub).
All of this LIVE!

Digital twins provide a powerful framework for creating high-fidelity virtual representations of real environments, enabling new forms of analysis, interaction, and decision-making. This talk will introduce the key principles behind building robust digital twins from real-world data, including capture workflows, reconstruction methods, and scalability considerations. I will present several use cases across domains such as cultural heritage, industrial environments and operations, virtual production, and interactive or gamified applications, illustrating how digital twins deliver measurable value in diverse contexts. The session will highlight how these models can integrate with computational pipelines, opening the door to applications that require high realism and scientific reliability. This general overview will establish a foundation that accommodates deeper exploration of specific applications during the live presentation.

Most computer-vision failures come from the rare cases, the dark corners, odd combinations, and edge conditions we never capture enough in real datasets. In this session, we walk through a practical end-to-end pipeline for generating targeted synthetic data using Google’s Nano Banana Pro and managing it with FiftyOne. We’ll explore how to translate dataset gaps into generation prompts, create thousands of high-quality synthetic images, automatically enrich them with metadata, and bring everything into FiftyOne for inspection, filtering, and validation. By the end, you’ll understand how to build a repeatable synthetic-first workflow that closes real vision gaps and improves model performance on the scenarios that matter most.