Yuning He Email & Phone Number

Mountain View, CA, US

• Tech Lead of Cloud Billing Reporting team. Designed and implemented the invoicing and reporting platforms to process tens of billions of usage records daily from millions of customers using Google Cloud, Workspace and.
• Project Lead in Cloud PA to migrate Google Cloud Marketplace business from "merchant of record" model to "agency of record" model to reduce partner frictions and regulatory risks. Responsible for aligning OKRs across.
• Designed and Implemented the Cloud Chargeback reporter to accurately recuperate cost from Google internal usages and meter the at-cost price from external usages, which enables rich P&L analysis on different level of.
• Designed and implemented disbursement pipelines for third-party partners to register and utilize Google Cloud Marketplace. Automated millions of dollars monthly revenue disbursement process with easier product.

Seattle, WA, US

• Designed and implemented a cluster level auto-scaling strategy to achieve better scalability over hundreds of terabyte data. Reduced the risk and blast radius of single cluster failure by 90% and lowered the cost of.
• Reserved Instance Recommendation: Implemented offline data pipeline to generate personalized Reserved Instance purchase recommendations based on customers’ historical on-demand usages and anticipated term of use..
• Cost Explorer Public API: Integrated Cost Explorer APIs into AWS SDK for getting cost and usage with compound filters and aggregations, getting EC2 Reserved Instances’ utilization, coverage and recommendations.
• Data Ingestion Cache Layer: Built a Memcached-based distributed two-layer cache to reduce the number of API calls to external services. Mitigated the risk of being throttled.

Pittsburgh, PA, US

I am a TA for 15-826 Multimedia Databases and Data Mining.Course web page: http://www.cs.cmu.edu/~christos/courses/826.F14/

Bedford, MA, US

1. Both sync and async performance benchmark about multi-clients connection to VoltDB.2. JUnit tests that cover all DDL features in VoltDB.3. Provide downloadable DDL from compiled catalog in VoltDB web studio via HTTP.4. Migrated VoltDB test infrastructures from using deprecated project.xml file to using pure DDL files5. Research on Jepson system that.

北京, Beijing, CN

1. Focused on key techniques for large scale data management.2. Implemented a distributed SQL query system on top of Hadoop. Replaced reduce phase with memory table opera-tions to do aggregations. Built a dynamic selector with heuristic rules and cost model.3. Experiments on Hadoop clusters show our system is four times faster than native Hadoop on average.

Armonk, New York, NY, US

I was in the WebSphere Team, focused on IBM Docs Viewer project.

Master'S Degree, Very Large Information Systems

Bachelor Of Science (B.S.), Computer Science

Bachelor Of Science (B.S.), Economics