Greater Madrid Metropolitan Area
Senior AI Engineer and data scientist building LLM-powered data products at Graphext. I design and develop structured output pipelines, prompt engineering frameworks, and natural language interfaces for data analysis — from Python and TypeScript libraries to full-stack AI applications. My background spans 15+ years across AI and machine learning: from computational neuroscience and evolutionary optimization of neural networks in academia, to real-time character AI at NaturalMotion (shipped in Red Dead Redemption, GTA), to leading ML engineering and data science teams. I bring deep experience across the full AI stack — NLP, generative AI, data pipelines, and production ML systems.
Building LLM-powered data analysis products with structured output and prompt engineering. Designed and developed Cuery (Python) and CueryJS (TypeScript) — libraries for structured LLM prompting over tabular data using OpenAI and Anthropic APIs. Integrated AI-driven insights, natural language querying, and automated enrichment into DatoCat and DatoFlow.
Led data science and ML engineering efforts. Built ML pipelines for intelligent data ingestion, automated feature engineering, and predictive modelling. Provided scientific and engineering support for an AI-guided data exploration platform.
Senior data science consultant working with various mobile network operators. Managed a small team responsible for the monitoring, analyzing and optimizing of network performance and its impact on customer experience.
I study social and biological networks and their dynamics. I'm currently investigating community dynamics in social networks (such as political affiliation and influence on twitter, currently in context of the Spanish general elections). I've also created the Worm Circuits Explorer, an interactive tool for exploration of the connectome of C. elegans.
I'm providing data analysis services to the flight meta-search engine voopter.com.br, involving e.g. the creation of an interactive dashboard (R/shiny) for exploratory statistics of flight searches (supporting identification of travel trends and recommendations).
Based in the University of the Basque Country and part of the European project "Extending Sensorimotor Contingencies to Cognition". Here I use evolutionary algorithms to optimize neural networks that solve cognitively interesting problems.