CDS Welcomes Plated’s Data Science Team

At the most recent Company Information Session, Plated explained how Data Science is shaking up the highly competitive meal-kit industry

NYU Center for Data Science
Center for Data Science
2 min readOct 10, 2017

--

At the most recent Company Information Session, CDS welcomed Plated, a meal-kit company led by chefs and data scientists.

Like most meal-kit companies, Plated offers a weekly meal subscription plan. Unlike other companies, however, they offer 20 different recipes per week (with dessert options!) and boasts a 100% turnover rate. This means that all 20 recipes change every single week. What’s the secret sauce to their success?

Data, of course. For example, let’s take Plated’s recommendation system. The typical data scientist’s gut instinct may be to cluster recipes together by protein, such as chicken. But this would cause the algorithm to continually recommend chicken dishes to the customer. Chicken’s great — but not every day!

The trick is not to have a one-size-fits-all recommendation system that is so commonly used in other industries.

When it comes to food recommendation systems, one needs to be more imaginative and wide-ranging: you need multiple models that cluster recipes according to different metrics (like calories, cuisine type, prep type, or color) so the algorithm can hit that sweet spot between consistency and variety.

Having recently been acquired by grocery company giant Albertsons, Plated is looking to expand their data science team to serve their rapidly expanding customers. Can our data students rise to the challenge? (The answer is yes, by the way.)

by Cherrie Kwok

--

--

NYU Center for Data Science
Center for Data Science

Official account of the Center for Data Science at NYU, home of the Undergraduate, Master’s, and Ph.D. programs in Data Science.