Feature Labs launched out of MIT to accelerate the development of machine learning algorithms by automating feature engineering. The company uses a method called Deep Feature Synthesis to create features from raw relational and transactional datasets, such as website visits or abandoned shopping-cart events. It publishes an open-source framework, Featuretools, to let developers experiment with automated feature engineering. For production and scaled projects it offers a commercial product available as a cloud service or on‑premises depending on customer requirements. Early customers cited in the article include BBVA Bank, Kohl’s, NASA, and DARPA. Feature Labs’ technology grew out of research at MIT CSAIL by founders Max Kanter, Kalyan Veeramachaneni, and Ben Schreck, with the idea beginning to form in 2015, and the company announced a $1.5M seed funding round that closed last March.

Total raised

$2M

Funding rounds

1

Latest round

Seed

Latest activity

Feb 2018