Built and leveraged an online homefeed recommendations platform to improve personalized content and increase engagement by +10%.

Experimented with content recommendations methods to optimize engagement (Saves, Clicks), freshness (age of images), localness (country and language), and other metrics related to a strong content ecosystem.

Leveraged Pinterest recommendations systems to intentionally distribute pins from the best content creators and reward users who produce engaging content for their audiences on Pinterest.

Instructed a section of 10-12 students in introductory CS class by leading students through weekly practice problems.

Graded around 25+ student submitted assignments each week for functionality and style by reading code and running tests.

Worked closely with professors Phil Levis and Nick McKeown to design class course material and assist 150+ students with Computer Networking projects.

Implemented the Commerce Merchant Catalog Viewer, giving merchants such as Nordstrom, Macys, and Shopify a self service tool to access Pinterest buyable pin product details.

Built out logic layers to surface 600,000+ buyable products as the first product API endpoints for all of Pinterest to leverage and build on .

Exposed the merchant catalogs further through a web UI on top of the API endpoints, making critical debugging information available to non- technical teams at Pinterest and partner companies.

Collaborated with the Cloud Datastore team to design and implement unindexed builtin properties, the #2 user requested feature on Datastore.

Designed and proposed methods of exposing this feature to over 250,000 active Google App Engine users.

Coded the backend implementation of this feature by working with the query planner, Megastore, and admin indexer, resulting in user savings of 12 cents per write for each property unindexed.