ShopRunner Processes 1TB of Data a Day to Recommend the Right Products

ShopRunner ingests product, logistic, and behavioral data from more than 100 retailers to provide highly personalized and customized experiences for its shoppers. It needs high-quality, fast, and efficient data management.

To automate its data pipelines, ShopRunner uses the Databricks Unified Analytics Platform on Amazon Web Services (AWS). Now ShopRunner improves its product recommendations due to better machine learning results.

In this webinar, you’ll see how ShopRunner:

• Created a product recommendation model using Apache Spark, as well as other machine learning frameworks like TensorFlow, on Databricks
• Shares custom libraries and notebooks for better collaboration between data engineers and data scientists, enabling smooth and automated data processes and accelerating innovation
• Gains greater insights with feedback from customer selections and results
• Ingests raw data from structured and unstructured file types provided by retailers - as much as 1 terabyte daily - and makes it immediately part of the website product recommendations