Tech lead manager for Shorts Platform Infra + Resource Efficiency

2016 Feb - 2020 Sep

Machine learning engineers to improve user experience with ads for Display ads quality

2020 Oct

Youtube Shorts

• Built map reduce pipelines to re-review the potential violation of keyword-inserted ads against Google policies

• Joined two very large datasets (several TB) and sent 10M ads for re-review in the pipeline to analyze potential violations

• Manufactured simulated large scale data with patterns on Hadoop Distributed File Systems to build proof of concepts

• Mined patterns from clients’ data in several industries and proposed predictive models for clients

• Researched on topics of “Large-scale malware behavior clustering on Hadoop”

• Experimented different Paralleled machine learning methods, including K-means, Canopy clustering, and Latent Dirichlet Allocation (LDA), with Mahout on Hadoop to leverage the framework for clustering big malware data

• Skills: Hadoop, Mahout