Akshay Agrawal

To power AI-driven customer insights, Zendesk needed to unify product telemetry and enterprise data. Legacy pipelines meant a single connector could take months to deliver value. By replacing a homegrown Airflow and custom connector stack with a Snowflake-powered data platform, Zendesk dramatically accelerated delivery. Insights that once took months now take weeks. In this session, learn how Zendesk achieved a 4x increase in connector output and scaled to more than 30 data sources with a lean team. We’ll explore the shift from siloed systems to a unified semantic layer — and how to prepare your data stack for custom SDK connectors, automated transformations and unstructured data.

Python notebooks are a workhorse of scientific computing. But traditional notebooks have problems — they suffer from a reproducibility crisis; they are difficult to use with interactive widgets; their file format does not play well with Git; and they aren't reusable like regular Python scripts or modules.