talk-data.com

Grammarly helps 30 million people and 50,000 teams to communicate more effectively. Using the Databricks Lakehouse Platform, we can rapidly ingest, transform, aggregate, and query complex data sets from an ecosystem of sources, all governed by Unity Catalog. This session will overview Grammarly’s data platform and the decisions that shaped the implementation. We will dive deep into some architectural challenges the Grammarly Data Platform team overcame as we developed a self-service framework for incremental event processing.

Our investment in the lakehouse and Unity Catalog has dramatically improved the speed of our data value chain: making 5 billion events (ingested, aggregated, de-identified, and governed) available to stakeholders (data scientists, business analysts, sales, marketing) and downstream services (feature store, reporting/dashboards, customer support, operations) available within 15. As a result, we have improved our query cost performance (110% faster at 10% the cost) compared to our legacy system on AWS EMR.

I will share architecture diagrams, their implications at scale, code samples, and problems solved and to be solved in a technology-focused discussion about Grammarly’s iterative lakehouse data platform.

Talk by: Faraz Yasrobi and Christopher Locklin

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

💡 Join our Data Platform team on June 21 to hear about Grammarly’s experience in adopting the Lakehouse architecture and migrating our data to Databricks with a small team. We have a lot of great insights and stories to share! 🙌

✅ Registration: https://gram.ly/42rtycX

► Speakers: Christian Acuna, Software Engineer Faraz Yasrobi, Software Engineer Chris Locklin, Engineering Manager

► Date: Wednesday, June 21, 2023, 18:00

► Who will be interested: Software Engineers, Data Engineers, Data Infrastructure Engineers, Machine Learning Engineers

► Summary: In this talk, we will share Grammarly’s journey of adopting the Lakehouse architecture and migrating our data to Databricks with a small team. You will learn how this transformation enabled us to do more with our data, unlock new use cases, and accelerate innovation. Join us to gain insights into the benefits of leveraging the Lakehouse architecture and the impact it had on Grammarly’s data platform, fostering a data-driven culture and driving significant improvements in product development and customer value.