Yingjie

About Yingjie

Solid mathematics and statistics background. Passion in data science, machine learning, programming.Experiences in end-to-end machine learning systems with Python, Spark, Docker, AWS, JavaScript. Building personalized recommender system and search engine.

Experience

1. Senior Machine Learning Engineer

   Atlassian

   Apr 2024 — Present · Mountain View, CA, US

   Multi-Stage Recommender System Framework Co-drove the design and implementation of the Multi-Stage Recommender (MSR) framework, combining wide-and-deep model architecture with multi-task learning to enhance generalizability and scalability Led the development of MSR framework by introducing modularized SDK, YAML-based configuration and MLOps practices. Championed the adoption of MSR framework across 5 business use cases. Led a XFN pod of 7 engineers and PMs, launching 4 large-scale online experiments, contributed 42.8% business metric gain towards the team’s half-year objectives.2. Realtime Confluence Template Recommender System Launched 4 online experiments, driving a 16% increase in signup rate (8K annualized signups). First author of GrowthPersonalization: Content Recommendation pending patent (U.S. Patent Application No. 19/004,447) Designed and implemented ML modeling and serving architecture for 3 recommender systems, including sequential embedding, semantic match, matrix factorization models. Built real-time inference pipeline by deploying streaming feature service and in-memory vector store with SageMaker, reducing latency and improving scalability. Drove decision-making and fostered alignment among 8 developers across 4 teams. Facilitated API service orchestration, experiment design, QA testing and monitoring.

Education

• UC San Diego

  Master of Science - MS, Computational Science of Mathematics and Engineering

  2018 — 2020

• Xidian University

  Bachelor of Science - BS, Mathematics and Statistics

  2014 — 2018