Jiting Xu is a Machine Learning leader based in San Francisco with 9+ years of experience building fraud, risk, and search ML systems across enterprises like eBay, Airbnb, and DoorDash. He blends hands-on engineering and team leadership—designing feature engineering, preprocessing, training pipelines, and offline/online evaluation frameworks that power anti-abuse, payment risk, and user-identity initiatives. At Voltron Data he led development of IbisML to enable large-scale, multi-backend ML preprocessing (including GPU backends), and he contributes backend data-compatibility features to the popular Ibis open-source dataframe project. His work is notable for turning complex cross-system data challenges into scalable pipelines and practical production models, including device-cookie graphing and large-scale segmentation for behavior scoring. Comfortable across Spark/Scala, Python, Kafka, and modern LLM/ML tooling, he excels at bridging research-quality models and robust, monitored production systems. Colleagues describe him as a technical lead who elevates both code quality and ML operational maturity while driving high-impact fraud detection programs.
3 years of coding experience
9 years of employment as a software developer
Bachelor of Science (BS), Engineering/Industrial Management, Bachelor of Science (BS), Engineering/Industrial Management at Wuhan University
Master of Science (MS), Computer Science, Master of Science (MS), Computer Science at University of South Carolina-Columbia
Bachelor of Science (BS), civil engineering, Bachelor of Science (BS), civil engineering at China University of Geosciences
Contributions:32 reviews, 17 PRs, 65 comments in 7 months
Contributions summary:Jiting primarily contributes to the Ibis project by implementing features related to data type conversions and adding support for operations in different backends. They have focused on converting data types like money and small money to decimal datatype, and adding UUID operation support for multiple databases. Furthermore, they enabled the creation of local backends with empty URLs, and added a `describe` method for computing summary statistics on table expressions. Their work extends to the addition of tests for TPC-DS queries.
Contributions:4 PRs, 105 pushes, 18 branches in 4 months
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.