Eric Hartfield
ECE M.S. student at UT Austin focused on applied ML for real-world sensing and decision systems.
I build models and data pipelines that turn raw, noisy measurements into reliable decisions. My work spans transformer-based sequence modeling, time-frequency image classification, wireless sensing, radar analysis, and interpretability studies that probe how models use physical structure rather than dataset artifacts.
About Me
I'm an ECE M.S. student at UT Austin focused on applied machine learning for real-world data. My work sits in the full ML loop: building data pipelines, designing representations, training PyTorch models, evaluating generalization, and analyzing how models make decisions.
I've worked across wireless sensing, radar, EEG, and graph learning, with recent projects using transformer-based sequence models, CNNs over time-frequency representations, and simulation-based augmentation. These projects have given me experience with messy datasets, distribution shift, feature extraction, model fusion, and experimental evaluation beyond clean benchmark accuracy.
I'm especially interested in ML systems where modeling and engineering both matter: systems that learn from complex data, expose meaningful structure, and remain reliable when conditions change.
Get In Touch
I'm interested in ML engineering, applied scientist, and software R&D roles where models, data, and evaluation all matter. My work focuses on building systems that learn from noisy real-world data, expose meaningful structure, and make reliable decisions under changing conditions.
If you're working on applied ML, model reliability, empirical evaluation, or research-driven software systems, I'd love to connect.