Data & Ml Engineer - Data & Ml @ Grubhub
Current• Build the pipeline and infrastructure to retain hundred millions of users per day for Grubhub CRM focuses. Transform strategy to execution. • Develop automated and distributed pipeline for leveraging near real time ML for precision targets and push marketing operation to excellence. • Developing full stack data science platform solution with s3, Data Lake, Hive, Presto to merge logistic and acquisition for delivering operation with CRM, SEM, paid social utm attribution for marketing and campaign planning and strategy across multi channels, region-markets, and canvas (Braze).• Drive sizable growth by launching experimental content and analytical metrics for branding and email/push notification programmatic campaigns with data engineering backend pipeline. • Integrating Machine Learning prediction (XGBoost, Time Series) and building ETL, feature processing, tuning, validation pipeline for implementing marketing conversions for growing Grubhub's drivers, diners, enterprise customers, and restaurants. • Build up Presto, Hive, Spark in Hadoop with Azkaban Scheduler to integrate debris operational, marketing, and finance data for Grubhub delivery decision.• Design and build data model and data structure in ETL and ELT based on pyspark, DataFrame, RDD.• Improve pipelines for analytical engineering and ML engineering and be agile with github, jenkins, and DevOps environment. • Predict CRM and Promo spending performance by training xgboost model in scikit-learn structure with marketing planning, targeting, creative features. Build up ML pipeline from data engineering, feature engineering, ML training and validation, model selection to web-based interactive app.• Integrate PCA + a/b testing to identify customers with profitable and maintainable ML model by using distributed pipeline from pyspark + azkaban, and implementing h2o package based on AWS EC2.
