Digital experiences at the intersection of AI, engineering and creativity.
I lead technology innovation by day and run a local AI lab by night: open-weight models on my own GPUs, real-time 3D and the occasional video game. This site collects what I learn along the way.
Local AI Lab
Open-weight models running on hardware I own: image and video generation, local LLMs and private RAG. Everything here was produced offline, on an RTX 3090 or an Apple Silicon Mac.
Local AI Lab
Local image & video generation with FLUX and Wan2.1
From DALL·E prompts in 2023 to a fully offline pipeline on an RTX 3090
Local AI Lab
Local LLMs: what fits in 16 GB vs 32 GB+
Gemma 4 12B on a laptop against Qwen3.8-27B on an RTX 3090 — measured, not guessed
Local AI Lab
Private RAG on your own machine with turbovec
Search 177K chunks of the Spanish official gazette locally — no cloud, no API keys
