Dylan Le is an AI/ML Engineer at Pangea, where he develops AI verification systems for reliability assessment in cloud environments. He also focuses on model compression techniques to reduce model size while preserving performance and robustness.
Dylan holds a B.S. in Computer Science and Mathematics and an M.S. in Computer Science from Rensselaer Polytechnic Institute. His academic and technical background spans machine learning, algorithms, distributed systems, and software verification. He has gained experience across both industry and research environments, including roles at Lockheed Martin and the U.S. Air Force Research Laboratory, where he worked on high-performance computing, computer vision, and applied AI systems.
At Pangea, Dylan contributes to the development of a peer-to-peer federated learning system integrated with homomorphic encryption, enabling secure, privacy-preserving collaboration across distributed data sources. He focuses on global AI model verification, ensuring consistency, robustness, and reliability of models deployed across decentralized environments. In parallel, he explores advanced model optimization techniques, including high-order SVD, to compress large-scale AI models while maintaining performance.