Anuradha U.

About Anuradha U.

As a seasoned professional in the realm of Machine Learning, I specialize in crafting innovative solutions within the domains of Search and Recommender Systems, contributing to the enhancement of efficiency and quality in people\'s lives.Currently, I spearhead the vibrant Machine Learning team at Shipt, focusing on Search, Personalization, and Ads ranking. Our significant efforts revolve around creating a multi-stage ranking framework for Search and Personalization. We leverage Learning to Rank (LTR) and Contextual Bandits for this purpose, fine-tuning Language Model (LLM) to bolster vector search capabilities. Additionally, our team pioneers Deep Learning-driven models to craft impactful Recommendations and Personalization strategies, enhancing the customization of products and content.In addition, I take the helm of the Machine Learning Platform Team at Shipt, steering the course for Shipt\'s MLOps Platform. This platform plays a pivotal role in managing the end-to-end lifecycle of Shipt\'s Machine Learning Workflow. From providing support for compute environments and feature stores to model registry, inference, and online prediction services, as well as model monitoring, our MLOps Platform is instrumental in meeting the high-demanding availability and SLA requirements for Shipt\'s Search & Recommendation, Catalog and Classification, and Bundle Engine.Prior to joining Shipt, my journey included serving as the founding Director of the PowerSchool Machine Learning team. I orchestrated the development of its MLPlatform on AWS, championing a CI/CD Driven model training deployment pipeline that supported the entire spectrum of Data Science work, from inception to model deployment.In my tenure at Snagajob, I made significant contributions by implementing the first-ever Sponsored Search algorithm and crafting the Yield Optimization Framework. This groundbreaking framework transformed Snagajob into a two-sided marketplace. My role extended to optimizing Search Ranking algorithms through advanced techniques like multi-stage ranking, Recall Optimization via query expansion using LSTM, and Click Models. Subsequently, I focused on enhancing user engagement through the implementation of the Snagajob Job Recommender, leveraging Bayesian Personalized Ranking (BPR) and emphasizing diversity, serendipity, and novelty in both Recommender and Search Results.My academic journey includes being an alumnus of the GroupLens Research Lab at the University of Minnesota, where I had the privilege of being advised by the late Professor John Riedl and Professor Joe Konstan.

Experience

1. Director, Ai & Machine Learning

   The Washington Post

   Sep 2025 — Present · Washington, DC, US

   Leading The Washington Post\'s AI and ML Team responsible for Conversational Search, Personalization and Revenue Optimization. Our team is responsible for The Washington Post\'s State of the Art ML driven PayWall, Ask The Post and site wide personalization.

Education

• Stanford University

  Professional Certificate of Artificial Intelligent, Artificial Intelligence

  2020 — 2021

• University of Minnesota

  PhD, Computer Science

  2013 — 2015

• University of Minnesota

  Master’s Degree, Computer Science

  2011 — 2013

• University of Minnesota

  Bachelor’s Degree, Computer Science & Minor in Mathematics

  2001 — 2006

• University of St. Thomas

  Master’s Degree, Software Engineering & Data Mining

  2007 — 2009

• Royal College Colombo

  Advance Level (A/L)