Kin Kan
Principal Member Of Technical Staff at Salesforce
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKin Kan is a Principal Member of Technical Staff based in Sunnyvale with eight years of professional experience building production ML systems and recommendation engines at Salesforce, and a longer research and engineering background across LinkedIn and Yahoo!. He blends applied research and product delivery—leading Einstein Prediction and Recommendation builder efforts and contributing core ML enhancements to the popular TransmogrifAI AutoML library for Spark, including XGBoost model tuning and feature-engineering fixes. Kin’s work spans end-to-end pipelines from offline experiments and feature engineering to online A/B launches, with particular strength in model interpretability and scalable scoring. He holds advanced engineering degrees from The University of Hong Kong and a PhD in computer science from UC Riverside, reflecting deep technical rigor underlying his practical impact.
7 years of coding experience
7 years of employment as a software developer
The University of Hong Kong (HKU)PhD, Computer Science, PhD, Computer Science at University of California, Riverside
Github Skills (12)
xgboost10
machine-learning10
spark10
feature-engineering10
automl10
scala10
pipeline9
automated-machine-learning9
ai9
mlops8
transformers8
python4
Programming languages (3)
Github contributions (3)
salesforce/TransmogrifAI
Jun 2018 - Mar 2019
TransmogrifAI (pronounced trăns-mŏgˈrə-fī) is an AutoML library for building modular, reusable, strongly typed machine learning workflows on Apache Spark with minimal hand-tuning
Role in this project:
ML Engineer
ML EngineerContributions:5 commits, 4 PRs, 7 pushes in 9 months
Contributions summary:Kin contributed to the core machine learning aspects of the project, specifically focusing on the `transmogrifAI` library. Their commits involve modifying feature engineering steps, fixing model insights for XGBoost models, and adding default parameter grids for XGBoost models. The user has also addressed issues in local scoring related to MultiPickList features and has improved the output of the XGBoostClassificationModel. These contributions indicate a strong focus on model training and feature engineering within the AutoML framework.
pythonsalesforcetransformationstransmogrifyestimators
faros-ai/metabase
Aug 2023 - Nov 2024
The simplest, fastest way to get business intelligence and analytics to everyone in your company :yum:
Contributions:35 reviews, 35 PRs, 303 pushes in 1 year 3 months
data-analyticsanalyticssimplestsqlfastest
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.