Summary
Jiayu Li is an applied scientist with a decade of experience bridging research and production at the intersection of generative AI, reinforcement learning, and efficient graph-based models. Currently at AWS after an internship on Bedrock, Jiayu has a strong publication record and led patents and papers on generative AI while building practical systems like HetGPT for heterogeneous graph neural networks. With a PhD background from Syracuse, they combine deep theoretical knowledge—graph sparsification, representation learning, spatiotemporal robustness—with hands-on skills in Python, C++, Java and frameworks like PyTorch and TensorFlow. Jiayu’s work spans federated and reinforcement learning for time-series domains (power networks, traffic forecasting) and efficient DNN compression, reflecting a rare blend of applied research and systems implementation. Colleagues note an ability to communicate complex ideas to broad audiences and to translate novel algorithms into production-ready components.
10 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD Computer and Information Sciences, Doctor of Philosophy - PhD Computer and Information Sciences at Syracuse University
Chinese, English, Chinese