Nov 2023 – Present · 2 yrs 5 mos · San Francisco, California, United States

• I am currently the Head of the New Verticals AI org at DoorDash. It’s an Applied AI/ML org that specializes in using machine learning at scale to all aspects of the the new verticals business at DoorDash. This includes:
• AI Agent Development
• Personalization & Discovery
• Search
• Consumer and Merchant Growth
• Order Fulfillment and Logistics
• Inventory Intelligence
• Computer Vision and VLMs
• ML driven Product Knowledge Graph & Catalog Building and Enrichment
• Taxonomy and Ontology

Aug 2022 – Nov 2023 · 1 yr 3 mos · San Francisco, California, United States

• I led the Personalization, Discovery and Search AI/ML team within the New Business Verticals ML Org at DoorDash.

Oct 2015 – Aug 2022 · 6 yrs 10 mos · San Francisco Bay Area

• As a Staff Engineer (L7) in Applied Machine Learning at Netflix, I lead cross functional projects in the algorithmic innovation space working closely with diverse teams.
• Over the years I have played a key role in identifying underutilized signals in Personalization and Search at Netflix, and driving cross team projects that led to algorithmic enhancements, and product feature launches resulting in measurable gains in member satisfaction.
• Some key highlights of my contributions at Netflix include improving the cold starting behavior of our main relevance ranker via a successful partnership with the metadata team, unlocking the value of the thumbs feedback signal to significantly improve personalization leading to broad impact all the way to UI/UX innovations, helping upgrade the search system at Netflix to a deep learning based model, developing the blueprint of a search intent classifier that is being used as a horizontal tool across many projects, and most recently, incubating a causal inference research pod that aims to unlock the incremental value of impressions.
• I am also passionate about data visualization, and championed a culture of visualizing the often opaque aspects of ML development cycle, such as feature importances, feature distributions, offline metrics, and input data patterns.
• I am an effective public speaker, and have represented Netflix in numerous external conferences, both technical and non-technical. Internally, I have evangelized ML to partner teams, which led to improved collaboration.

Feb 2014 – Oct 2015 · 1 yr 8 mos · San Francisco Bay Area

• I am passionate about using data science techniques to enable a personalized dining experience. Current projects in my team include applications of natural language processing and topic modeling to our extensive review corpus to reveal salient features of restaurants and reviewers. I am also helping develop an extensive recommendation systems stack using user interactions and restaurant metadata.
• We are hiring!

Nov 2013 – Jan 2014 · 2 mos · San Francisco Bay Area

• Bringing contextual music recommendations to life!

Aug 2012 – Nov 2013 · 1 yr 3 mos

• Analysis and interpretation of massive datasets from various telescopes, looking for correlations and patterns. Current projects rely on Machine Learning techniques: Bayesian inference, Gaussian processes, MCMC, model selection, and regression analysis.

Sep 2009 – Aug 2012 · 2 yrs 11 mos

• Led a team to isolate subtle signals in the massive dataset from the ACT telescope, and made the first ever detection of an elusive astrophysical effect. Led the effort to estimate the power spectrum of cosmic radiation, and Bayesian estimation of model parameters.

Nov 2008 – Sep 2009 · 10 mos

• Cosmology Data Analysis Postdoctoral Fellow

Sep 2004 – Nov 2008 · 4 yrs 2 mos

• Led the development of a Python Scipy Numpy based framework for analyzing Cosmic Microwave Background data.