Bio & Trajectory

Biography

I am a Research Scientist at Tecnalia Research & Innovation (Derio, Bizkaia, Spain) and a PhD Candidate in Energy Engineering and Artificial Intelligence at the University of the Basque Country (UPV/EHU). My doctoral dissertation is co-supervised by Dr. Roberto Garay Martinez (Principal Investigator of Energy at DeustoTech, University of Deusto) and Dr. Iñigo Mendialdua (Department of Computer Languages and Systems, University of the Basque Country).

My research lies at the convergence of Artificial Intelligence, Deep Reinforcement Learning (DRL), Model Predictive Control (MPC), and Thermal Building Physics. Heating, ventilation, and air-conditioning (HVAC) systems constitute one of the largest consumers of primary energy in the European Union. My work focuses on designing intelligent supervisory control operators capable of anticipating variable weather conditions, fluctuating electricity prices, and building thermal inertia to minimize operational costs and carbon emissions while guaranteeing indoor occupant comfort.

Empirical Research at KUBIK Testbed

A distinctive pillar of my doctoral work is empirical validation on full-scale physical infrastructure. Our control policies—formulated with algorithmic paradigms including Deep Q-Networks (DQN), Proximal Policy Optimization (PPO), and Deep Deterministic Policy Gradient (DDPG)—are trained against calibrated transient building simulations and deployed onto Tecnalia’s KUBIK experimental test facility, a unique multi-zone smart building capable of real-world energy and climate testing.

Trajectory & Academic Background

I earned my Bachelor's Degree in Computer Engineering and Information Systems from the University of the Basque Country (UPV/EHU). Prior to joining Tecnalia in 2019, I worked as a Technical Researcher at the University of Deusto (MORElab / DeustoTech) in Bilbao. There, I collaborated on national and European research initiatives (including the Horizon 2020 City4Age project), investigating Internet of Things (IoT) middleware architectures, Linked Open Data semantics, and machine learning models for Ambient Assisted Living (AAL) and smart cities.

Since joining Tecnalia, I have contributed to competitive European and national projects in energy systems and automation, including the newly awarded Horizon Europe EIC Pathfinder Challenge project SITEBOT (Swarm robotic Intelligence for TimbEr Building On-site Technologies, Grant Agreement #101307035, 2026–2030).

Core Research Interests

• Deep Reinforcement Learning for Physical Systems: Generalized reward shaping, safe exploration, and sample-efficient policy optimization in cyber-physical energy systems.
• Model Predictive Control (MPC): Multi-horizon predictive scheduling, evolutionary optimization, and hybrid physics-informed neural network controllers.
• Building Energy Management (BEMS): HVAC modulation, Ground Source Heat Pumps (GSHP), thermal energy storage, dynamic electricity tariffs, and district heating networks.
• Digital Twins & Real-World Testing: High-fidelity transient building models (Modelica, EnergyPlus), sensor analytics, and experimental validation at KUBIK.