Peter Wang
Machine Learning Engineer at Meta
San Jose, California, United States
Summary
🤩
Rockstar🎓
Top SchoolPeter Wang is a Machine Learning Engineer with 7+ years building production-grade ML systems and a deep background in large-scale recommendation, streaming, and big data infrastructure. He designed and owned recommendation platforms at Roblox—authoring both batch and streaming personalization pipelines—and previously scaled recommenders, graph databases, and A/B tests at Smule. His skill set spans Spark, AWS, real-time pipelines, ensemble and classical ML (GLM, CART, PCA, clustering), geospatial smoothing/kriging, and survival/network linkage analyses, with R Shiny experience for rapid prototyping. Now at Meta after leading ML platform efforts and supervising engineers, he brings product-minded architecture experience alongside hands-on implementation. An active contributor to DataHub, he improved Superset ingestion, column-level lineage, and robust API handling to reduce hanging queries—demonstrating an attention to data catalog reliability as well as model quality. Based in San Jose, he pairs applied-math training from UIUC with pragmatic engineering to move complex ML from research into resilient production.
7 years of coding experience
16 years of employment as a software developer
Bachelor's Civil Engineering, Bachelor's Civil Engineering at Tongji UniversityMaster's Applied Mathematics, Master's Applied Mathematics at University of Illinois Urbana-Champaign
English
Github Skills (14)
data-ingestion10
data-lineage10
discovery10
data-governance10
data-catalog10
metadata10
service-discovery10
auto-discovery10
python10
apidoc9
api9
multithreading8
github-ci4
githubaction-workflow4
Programming languages (8)
Github contributions (5)
datahub-project/datahub
Jan 2025 - Mar 2025
The Metadata Platform for your Data and AI Stack
Role in this project:
Back-end Developer
Back-end DeveloperContributions:20 reviews, 8 PRs, 30 comments in 2 months
Contributions summary:Peter's contributions center around enhancing the Superset data ingestion capabilities within the DataHub project. They implemented features to integrate Superset's dataset lineage, and added ownership information for charts, dashboards and datasets, improving the data catalog. Furthermore, the user introduced column-level lineage for both datasets and charts, augmenting the data lineage capabilities. They also addressed potential issues with hanging queries and resource usage by introducing timeout values to the Superset API calls and leveraging threads for API calls, improving system performance.
data-managementdata-discoverydata-stackmodern-data-stackdata-catalog
PeteMango/PeteMango
Dec 2022 - Feb 2025
Contributions:2 PRs, 57 pushes, 1 branch in 2 years 2 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.