Pinterest is building a world-class shopping experience for our users; one of our top user requests is to be able to purchase the cool things they find on Pinterest. The Shopping Content Mining team in Toronto plays a critical role in this journey. Each Pin at Pinterest links to a URL, this team extracts helpful data like product prices and recipe ingredients from the HTML. The team manages the entire ML lifecycle, from labeling systems to model development to serving and measuring data quality ops. Our team consists of backend, frontend and ML engineers, along with a team of contractors to do labelling.

The team’s ML Engineers research and develop ML systems to interpret webpages. This is a unique problem space with lots of possibilities for solutions including graph neural networks, NLP, computer vision and simple linear models. Our models are used to detect erroneous data, classify webpage types, and extract information. ML Engineers on our teams develop end to end solutions, from data gathering to model training and productionization.

What you’ll do

• Design and build large scale ML systems that can process billions of webpages
• Train ML models to create representations of HTML pages, classify webpages and extract structured information from webpages.
• Create NLP systems to extract information from free form text and link
• Manage lifecycle and productionization of ML solutions, from data gathering to data quality management to measuring impact via experiments
• Drive cross-functional collaborations with partner teams
• Help research novel solutions and present results via papers, blog posts or conference talks

What we’re looking for:

• 2+ years of industry experience
• Hands-on experience on large scale machine learning systems (full ML stack from modeling to deployment at scale.)
• Experience with production large scale serving or data systems

Nice to have:

• PhD/Masters in Machine Learning or related areas, publication on top ML conferences
• Familiarity with information extraction techniques for web-pages and free-texts
• Experience working with shopping data is a plus

This position is not eligible for relocation assistance.

#LI-HYBRID  #LI-MF1