Sheng Chen is a seasoned software engineer and machine learning specialist with 11 years of experience building production ML systems and backend services, currently at Google in New York. He holds a PhD in machine learning and a strong publication record from Stevens Institute of Technology, where he developed adaptive incremental and imbalanced learning models for streaming and skewed-data problems. Sheng has productionized ML at scale—improving record linkage and active learning pipelines at Intelius and designing ranking and non-blocking web services at Demand Media—and served nearly nine years in engineering and VP roles at Bank of America. A polyglot programmer comfortable in Python, Java, C and Scala, he is an active open-source contributor to prominent Scala projects like fs2 and http4s, enhancing stream processing and WebSocket/http utilities. Notably, his academic work on convergence and error bounds for streaming learners underpins practical systems that handle skewed, real-world datasets.
11 years of coding experience
6 years of employment as a software developer
B.Sc., Electrical Engineering, B.Sc., Electrical Engineering at Huazhong University of Science and Technology
Ph. D., Computer Engineering / Machine Learning, Ph. D., Computer Engineering / Machine Learning at Stevens Institute of Technology
Contributions:10 commits, 5 PRs, 1 push in 7 months
Contributions summary:Sheng primarily contributed to the `blaze-core` module, likely focusing on WebSocket-related functionalities within the http4s library. Their work involved modifying the `Http4sWSStage.scala` file to address issues and improve its internal logic. The user also introduced enhancements to the `dsl` package, particularly the `Method` class, implementing an "or" extractor. Furthermore, the user added utility functions for building `UriForm` objects.
Contributions:10 commits, 5 PRs, 16 comments in 3 days
Contributions summary:Sheng made significant contributions to the `fs2` library, a compositional, streaming I/O library for Scala. Their work involved implementing new functionalities for `Process1`, including `takeWhile`, `dropWhile`, and `zipWithIndex`, and refactoring existing methods. They also updated and expanded the testing suite to cover the new features and refactored code to preserve the chunkiness of the stream. These changes directly enhance the library's capabilities for stream processing.
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.