Machine Learning: improve automated cross-platform rebudgeting service to automatically focus spend on the best ROI/CAC campaigns. Focus on reliability & correctness to allow for faster & more confident A/B testing of novel rebudgeting strategies for all of Netflix’s programmatic ad spend.

Machine Learning: implementing new machine learning features (1% - 5% improvements in target metrics). Hold Patent Application No. 16/023,894 for contextual search suggestions to interactively improve the query; and Patent Application No. 16/446,364 for contextual search ranking using entity topic representations

Diversity Relevance: architecting and implementing representative reranking, resulting in 95% of all searches being gender-representative with respect to the qualified population of the search (https://lnkd.in/gBPmGxe)

Backend Engineering: improving performance to significantly reduce operating expenses (substantially reduced the number of servers needed)

Full Stack work: growth - increase revenue 5 - 15% on different subsets of our advertiser base with each successful experiment by improving our ads creation experience; frontend - working on creating the new delivery insights product; backend - scaling backend QPS, adding new features and data pipelines

Worked with many disparate parts of the codebase on allowing developers on the Parse platform to implement arbitrary authentication, including webhook verification

Backend (primarily): Go, C++, Ruby

Frontend: Ruby on Rails, HTML 5/CSS 3/Javascript

Mobile: Objective C (iOS), Java (Android)