Alexa Audio Experience

Twitch Monetization

- established ML practices in both Commerce and Ads- close collaboration with DS, Economists, Product and Engineers to deliver complex projects Monetization:- Allocation of community gifts to maximize conversion - Emote embedding for large scale similarity search- Personalized Ad running configurations with revenue forecasting to incentivize participation - Live contextual scheduling of video ads during live streams to reduce viewer impact

We’re on a mission to deliver life changing opportunities by removing obstacles that stifle employment prospects, to help people and businesses grow. Powered by unique AI technology, Uncommon IQ™ is the industry’s first fully-automated, transparent and objective recruiting platform that connects companies with only interested and qualified candidates.Lead a team of Machine Learning Engineers and company strategy around machine learning and data science:- Responsible for hiring, performance management and career growth- Manged all aspects of product development including prioritization, estimation and architecture- Built qualification based job board bidding system based on Bayesian forecasting and optimization- Built AI system for automated qualification of inbound job applicants- Built large scale (> 100 M ) passive search product powered by machine learning- Built system for end to end automated sourcing

First Data Scientist, built entire DS/ML stack from scratch- Architected and built Docker based micro-services data engineering pipeline to ingest application events for machine learning- Built hybrid item co-occurrence and latent user feature based system to recommend items to stylists for box curation that outperformed unaided human stylist selections-Built system to mine free form customer feedback for product requests using semantic product search backed by NLP and ElasticSearch- Built product attribute based recommenders to personalize box curation and increase wishlist engagement- Scraped 10s of competitors websites and used a deep neural network for image analysis of competitor products

Implemented predictive modeling based on large-scale data to score and stack rank developers for Gild Source.

Prof. Han Lim, Integrative BiologyInstructed students in the use of MATLAB for modeling and simulation of biological systems and statistical analysis of simulation data. Developed rubrics, led review sessions, and assessed student progress for four undergraduate courses. Oversaw a multi-week special project where students pose and answer individual research questions in systems biology. Fall 2007 Introduction to Protein InformaticsProf. Kimmen SjoanderDeveloped computer lab exercises for computational phylogenetic analysis of protein families for structure and function prediction. Lead two computer lab sections of 30 advanced undergraduate students and instructed them in the use of computational tools for biological sequence analysis including multiple sequence alignment, construction of hidden Markov models (HMM) and phylogenetic tree construction. Developed and graded tests and assessed final projects where students performed novel phylogenetic analysis.

Worked on IC biosensor development for point-of-care immunological assays.Included micro-fabrication and assay development.