Summary
Yang Bai is a research scientist and Ph.D. candidate at the University of Florida with nine years of experience building representation learning and retrieval-augmented systems for NLP, IR, and multi-modal generation. Now at Meta working on Instagram Ads and previously on Threads search, he bridges cutting-edge research with product-focused engineering, having shipped retrieval and ranking models in industry internships. His academic work—supported by DARPA funding—includes a SIGIR2023 multi-answer QA benchmark, novel evaluation metrics, and leading annotation efforts and tooling for controversial stance and check-worthiness detection. He has practical experience scaling dense retrieval (e.g., a log-retrieval system at Nokia Bell Labs that beat BM25 by 16.1%) and a background spanning Python, C/C++, Java, and both PyTorch and TensorFlow stacks. Comfortable leading small teams and building end-to-end pipelines, he combines deep academic rigor with an eye for customer-centric, scalable product design.
9 years of coding experience
7 years of employment as a software developer
Bachelor of Science - BS Microelectronics, Bachelor of Science - BS Microelectronics at Sichuan University
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at University of Florida
English, Chinese