Personalization:Architected and implemented modern machine learning and deep learning recommendation systems at scale to provide personalized recommendations to Target guests.Demand Forecasting:Built deep forecasting capabilities to solve various forecasting problems at massive scale. Enhanced the existing demand forecasting engine (DFE) to improve its coverage, scalability, reliability and efficiency.

My role is a hybrid of research, modeling and model implementation. I build machine learning models (deep learning, xgboost, SARIMAX, etc) and wrote product code for Grubhub delivery

• Owned, designed and deployed Gradient Boosted Trees models end-to-end to predict critical component failures on mining trucks using time series sensor data. Model Achieved ~7 days leading time with > 85% precision and recall.• Architected and built end-to-end automated ML engine to automate scientists’ work at scale: Built a vertical agnostic information filtering engine to detect and suggest co-occurring and abnormal patterns on discrete time series data. • Developed Airflow pipelines to conduct customer pilot, monitor model performance on real time data in product.• Frequently communicated and presented data insights and model key metrics through flask/shiny dashboard to senior leaders, clients and subject-matter experts.

Job title normalization and taxonomy.