Hao Xu
Senior Software Engineer, Machine Learning Infrastructure at Apple
California, United States
Summary
🤩
Rockstar🎓
Top SchoolHao Xu is a senior software engineer specializing in machine learning infrastructure and large-scale data systems, with nine years of experience building production ML and data platforms. He has led feature store and streaming/batch infrastructure work at companies like Affirm, Fast, JPMorgan Chase, and now Apple, and co-founded Aita to deliver cloud-native AI analysis tools. Hao has deep backend and distributed-systems expertise, having implemented vector-search support for Feast (integrating Elasticsearch and Postgres to enable embedding-based feature retrieval). His work spans the full ML lifecycle—feature engineering, serialization, online stores, and observability—while operationalizing pipelines for millions of domains. Based in California, he combines startup agility with enterprise-grade architecture and a knack for simplifying complex data workflows to reduce cost and improve reliability. An uncommon strength is his cross-cutting focus on both core infrastructure internals and practical integrations that make advanced ML features accessible to product teams.
9 years of coding experience
9 years of employment as a software developer
Bachelor's degree, Bachelor's degree at Shandong UniversityMaster’s Degree, Master’s Degree at Carnegie Mellon University
Github Skills (14)
elasticsearchquery10
elasticsearch10
vector-database10
feature-store10
postgresql10
aws-elasticsearch10
python10
elasticsearch-api10
amazon-elasticsearch10
data-engineering9
machine-learning8
data-science8
big-data7
go4
Programming languages (7)
Github contributions (5)
feast-dev/feast
Feb 2024 - Apr 2025
The Open Source Feature Store for AI/ML
Role in this project:
Back-end Engineer
Back-end EngineerContributions:169 reviews, 93 PRs, 150 pushes in 1 year 2 months
Contributions summary:Hao primarily focused on implementing and integrating vector database functionality, specifically for Elasticsearch and Postgres within the Feast framework. They added support for online document retrieval, enabling the use of embeddings for feature search. The user's work involved modifying core online store components and creating supporting infrastructure to handle vector similarity searches and feature value storage. The user also made improvements to serialization and deserialization of entity keys.
pythondata-qualitydata-sciencemlmachine-learning
project-ryoma/ryoma
Feb 2024 - Nov 2024
Common AI agent framework solving your data problems
Contributions:8 reviews, 123 PRs, 184 pushes in 8 months
aidata-sciencellm
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.