ML Engineer & Researcher | Multi-Agent Systems | Mechanistic Interpretability
Building production ML systems and researching how neural networks learn and coordinate. Interested in understanding model internals to build safer, more reliable AI.
MS Computer Science, University of New Mexico (2024)
- Mechanistic Interpretability — Understanding learned representations and decision-making in neural networks
- Multi-Agent Systems — Coordination, communication, and emergent behavior in RL agents
- Reinforcement Learning — Value estimation, exploration strategies, RLHF
PAW: A Deep Learning Model for Predicting Amplitude Windows in Seismic Signals
A.M. Villegas Suarez, D. Reiter, J. Rolfs, A. Mueen
DSAA — Paper
MatchMakerNet: Enabling Fragment Matching for Cultural Heritage Analysis
A.M. Villegas-Suarez, C. Lopez, I. Sipiran
ICCV Workshop on e-Heritage — Paper
Enhancing Value Estimation Policies by Post-Hoc Symmetry Exploitation in Motion Planning Tasks
Y. Hasan, A.M. Villegas-Suarez, E.C. Carter, A. Faust, L. Tapia
IROS — Paper
More in Google Scholar