Summary
Cat Le is an AI Scientist with a decade of experience developing task-aware machine learning methods spanning neural architecture search, few-shot and continual learning, generative image models, and NLP. She bridges deep academic research—authoring a novel task-affinity metric based on Fisher information and matching algorithms during her PhD—with applied R&D at industry leaders like Amazon, Raytheon, and UL Solutions. Her recent work focuses on AI safety, robustness, and standards alignment, including building LLM-powered evaluation tools to streamline ISO/IEC compliance workflows. Cat has a strong experimental track record in transformer and attentional-copula architectures for time-series, defect detection, and backscatter imaging, and she blends systems-level engineering (firmware and OCR for camera systems) with high-impact research. Based in Durham, NC, she pairs rigorous theoretical insight with practical productization, often integrating visual and textual modalities to solve domain-specific problems.
10 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Duke University
California Institute of Technology
Bachelor of Science - BS, Electrical and Computer Engineering, Bachelor of Science - BS, Electrical and Computer Engineering at Rutgers University