Ricky X is a Member of Technical Staff and systems engineer with 11 years of experience building high-performance distributed systems for ML and databases, currently working on training infrastructure at OpenAI. He previously drove core Ray runtime and LLM inference performance at Anyscale—contributing to vllm and Ray internals—and has hands-on database internals experience from CMU projects like Noisepage and BusTub. Comfortable across research and production, he focuses on performance optimization, profiling, and maintainable API design, often tackling low-level bottlenecks such as KV cache management and task failure observability. Based in Mountain View, he blends academic rigor (CMU MS) with industry impact and a habit of “simply enjoying complex systems,” evidenced by sustained open-source contributions to widely used projects in the ML infra ecosystem.
11 years of coding experience
5 years of employment as a software developer
Hangzhou Foreign Language School
High School Iternational Baccalaureate Diploma Programme, High School Iternational Baccalaureate Diploma Programme at Anglo-Chinese School (Independent)
Bachelor’s Degree Computer Science, Bachelor’s Degree Computer Science at National University of Singapore
Master of Science - MS Computer Science, Master of Science - MS Computer Science at Carnegie Mellon University
NDO Computer Science, NDO Computer Science at Stanford University
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Role in this project:
Back-end Developer & System Architect
Contributions:3 releases, 1338 reviews, 135 commits in 7 months
Contributions summary:Ricky focused on refactoring the state API server's interaction with the submission client. They created a `StateApiClient` that inherits from `SubmissionClient` and refactored various listing commands into class methods, demonstrating an understanding of API design and back-end architecture. The user also implemented a feature to track down the detailed reason for each failed task, including error type and message, contributing to the system's observability. Additionally, the user has fixed and corrected the code issues in different files to make it more robust.
Self-Driving Database Management System from Carnegie Mellon University
Role in this project:
Back-end Developer / Database Engineer
Contributions:50 reviews, 5 commits, 8 PRs in 2 months
Contributions summary:Ricky contributed significantly to the database management system's distinct aggregation functionality. They focused on implementing and refining the logic for distinct aggregations, including COUNT/MAX operations, and incorporating group by terms. The user's work involved creating and modifying filters, hash table implementations, and key structures essential for efficient distinct aggregation. Their contributions included fixing bugs, optimizing code, and adding trace information to the system.
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.