Yitong Xu is a Senior Staff Engineer with 12 years of experience building ML-powered search and recommendation systems, currently leading Agentic Evaluation pipelines and Harness Engineering at Baidu to make non-deterministic LLMs reliable for production autonomous agents. He has deep expertise in search ranking and foundation LLM training, previously designing scalable unsupervised models for query intent, keyphrase extraction, and churn prediction at Audible and Baidu. Yitong combines strong research chops from Columbia and Yale with hands-on production engineering—he has contributed to the nanomsg open-source library by improving core data structures, memory management, and build processes. Comfortable across the ML stack, he blends statistical rigor with software engineering discipline to ship high-impact systems at scale. An implicit strength is his knack for turning academic advances (adversarial training, variational models) into robust, deployable features that improve user-facing relevance.
12 years of coding experience
3 years of employment as a software developer
Master of Science (M.S.), Data Science, Master of Science (M.S.), Data Science at Columbia University in the City of New York
Contributions summary:Yitong's contributions primarily focused on improving the nanomsg library's core functionality and code quality. They implemented a hash table rehashing mechanism, fixed several bugs related to double inclusions and memory management, and refactored code for better style. The user also made changes related to the project's build process and general improvements to the code base, as well as removing unused parameters in different APIs.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.