Andres Felipe Posada Moreno

Postdoctoral Researcher — Foundation Models for Dynamical Systems & Explainable AI

Hi there! Since September 2024, I am a postdoctoral researcher at the Institute for Data Science in Mechanical Engineering (DSME) at RWTH Aachen University. I completed my PhD under the supervision of Sebastian Trimpe, and was an associate doctoral researcher at the Cluster of Excellence “Internet of Production” funded by the DFG.

Currently, I focus on Foundation Models for Dynamical Systems, exploring transformer-based models pretrained on synthetic data for prediction and control. This research is supported by the JUPITER AI Factory (JAIF), part of the EuroHPC initiative. I also continue my work on eXplainable Artificial Intelligence (XAI), developing methods to understand the representations learned by deep neural networks from images and time series.

Across these directions, I am interested in what neural networks learn, how they learn, and the capabilities and limitations of their representations. My goal is to use this understanding to help neural networks learn more effectively and at scale in useful domains.

Prior to that, I completed a bachelor’s degree in mechatronics engineering at the EIA University, a master’s degree in engineering at the Arts et Métiers ParisTech, worked on agroindustrial applications of IoT and AI at Asimetrix, and worked as a researcher on applied AI at the IMA Cybernetics Lab, RWTH Aachen University, in the context of the QUISS project.

Research Interests

Foundation Models for Dynamical Systems
Transformer-based models pretrained on synthetic data for prediction and control of dynamical systems.
Explainable Artificial Intelligence
Understanding the representations learned by deep neural networks and developing methods to explain their behavior.
Applied AI
Industrial applications of AI across manufacturing, quality control and transport.

Projects

JUPITER AI Factory (JAIF) EuroHPC AI Factory built around Europe’s first exascale supercomputer.
Internet of Production (IoP) DFG-funded excellence cluster on using AI in production engineering.
QUISS Using AI for damage recognition in rail transport.

Education

PhD in Artificial Intelligence — RWTH Aachen University, Germany (2024)
M.Sc in General Engineering, Mechatronics Expertise — Arts et Métiers ParisTech, France (2014)
B.Sc in Mechatronics Engineering — EIA University, Colombia (2012)

Selected Publications

CENDRe: Concept Extraction with Natural Domain Representations (2026)
Holzapfel, A., Posada Moreno, A.F., Trimpe, S. — NeurIPS 2026
Details | arXiv
The risk of KV cache compression (2026)
Haverbeck, L., Amo Alonso, C., Posada-Moreno, A.F., Trimpe, S., Pavone, M. — NeurIPS 2026
Details | arXiv
Sailing Towards Zero-Shot State Estimation using Foundation Models Combined with a UKF (2025)
Holtmann, T., Stenger, D., Posada-Moreno, A., Solowjow, F., Trimpe, S. — CDC 2025
Details | DOI
On Foundation Models for Dynamical Systems from Purely Synthetic Data (2024)
Ziegler, M., Posada-Moreno, A.F., Solowjow, F., Trimpe, S.
Details | arXiv
ECLAD: Extracting Concepts with Local Aggregated Descriptors (2024)
Posada-Moreno, A.F., Surya, N., Trimpe, S. — Pattern Recognition
Details | DOI

Contact

Email: andres.posada@dsme.rwth-aachen.de
Location: Aachen, NRW 52070, Germany
Google Scholar | GitHub | LinkedIn