Yucheng Low
Principal Engineer at AI2 Incubator
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolYucheng Low is a principal engineer and founder with 15+ years building high-performance ML infrastructure and distributed systems, grounded in a PhD in Machine Learning from Carnegie Mellon where he worked on GraphLab (a project that later earned a VLDB Test of Time Award). He co‑founded GraphLab/Dato/Turi and architected SFrame and other scalable, out‑of‑core analytics primitives, then led dataset storage and on‑device model conversion efforts after Turi was acquired by Apple. As CEO/cofounder of XetHub he tackled ML data management at scale and, after acquisition by Hugging Face, architected privacy-sensitive client-side deduplication that scales to tens of petabytes. Now based in Seattle, he continues to drive systems architecture at Qumulo and advises AI startups through AI2’s incubator. His contributions to PowerGraph and low-level aggregation primitives reflect deep expertise in threading safety, memory pools, and high-throughput backend design. Colleagues describe him as a rare blend of research pedigree, product instinct, and hands‑on systems craftsmanship.
15 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy (PhD) Machine Learning, Doctor of Philosophy (PhD) Machine Learning at Carnegie Mellon University
English
Github Skills (8)
distributed-systems10
c-language10
cprogramming-language10
multithreading10
template-metaprogramming9
algorithms8
algorithm8
software-design8
Programming languages (8)
Github contributions (5)
jegonzal/PowerGraph
Sep 2010 - Jun 2013
PowerGraph: A framework for large-scale machine learning and graph computation.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1696 commits, 2 PRs, 2 pushes in 2 years 8 months
Contributions summary:Yucheng's commits primarily focus on the implementation of aggregation functionality within the PowerGraph framework. They introduced a partial implementation of aggregation, including the `distributed_aggregator` class, with features such as vertex and edge aggregators. The user's work involved modifications to the header files, with a focus on threading safety, the addition of memory pools, and integrating this functionality with existing code for testing with the consistent engine.
computationcomputation-graphmachine-learningscalelarge-scale-machine-learning
ylow/turicreate
Dec 2017 - Nov 2024
Contributions:1 release, 1 review, 1 PR in 7 years
machinelearningmachine-learningmachine-learning-models
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.