Snowflake Ecosystem Podcast

S01E05 Data Observability with Soda

In episode 5, Koen Van Duyse (Soda.io) discusses data observability at scale with Joris Van den Borre (Tropos.io).

🚀 In this podcast, Koen Van Duyse (Soda) discusses data observability with Joris Van den Borre (Tropos.io). This emerging field in data management ensures continued trust in data pipelines at scale. Why Koen and Soda? Koen has been at the forefront of data governance and cataloging and since joining Soda, he is pioneering data observability with Soda today.

This episode of the Snowflake Ecosystem Podcast will show how data analytics leaders bridge the gap between technical data quality and business needs.

Episode highlights

Reimagining Data Quality: Soda is redefining the traditional notion of data quality and data governance, offering practical solutions based on LLM’s to decentralize data quality rule definition.

✅ Empowering Non-Technical Users: Explore how Soda’s platform solutions make data quality accessible to a broader audience, breaking down the barriers between technical and non-technical teams.

✅ SodaGPT: SodaGPT – a chatbot-based rule definitions engine – is pioneering human-readable and writable DQ language, ushering in a new era of data agreements.

Real-World Applications: Delve into the real-world applications of Soda’s technology and its seamless integration into modern data stacks, providing tangible benefits for organizations.

Get started with Soda for data observability in Snowflake: soda.io 

Scroll to Top