🛡️Preventing Broken Dashboards with Trino and dbt
🌍 Format: Online only
📌 Language: English
🔍 Event Description
Few things are more frustrating than opening a dashboard in Superset, Tableau, or Power BI only to discover that it no longer works because an upstream transformation model has been changed or deleted.
And often, the person making that change has no idea that a critical dashboard depends on it.
In this session, we'll explore how Trino and dbt can be used together to proactively identify and prevent these kinds of dependency issues before they reach production.
Drawing from the open-source project superset-dbt-trino-guard, we'll look at how metadata from dbt projects and BI dashboards can be combined to understand downstream dependencies, validate them with Trino, and integrate those checks directly into CI/CD pipelines.
The result is a practical approach to protecting dashboards, reducing unexpected downtime, and making changes to transformation models safer across the entire analytics platform.
The key areas we'll explore are:
Capturing metadata from dbt projects and BI dashboards
Understanding dependencies between transformation models and dashboards
Validating downstream dependencies with Trino
Detecting potentially breaking changes before deployment
Automating dependency validation within CI/CD pipelines
Building a reusable approach that works across multiple BI tools
👥 Who Should Join?
Data engineers and analytics engineers
dbt and Trino users
Superset, Tableau, and Power BI developers
Data platform engineers and architects
DevOps and CI/CD practitioners working with analytics platforms
Anyone interested in improving reliability and preventing breaking changes in modern data platforms
✨ What to Expect
A practical look at a common but often overlooked data platform problem
Live demonstrations based on the superset-dbt-trino-guard open-source project
Insights into capturing and combining metadata from dbt and BI platforms
A practical approach to validating dependencies before changes reach production
Guidance on integrating dependency checks into CI/CD workflows
A reusable pattern that can be applied beyond Superset to other BI tools
📍 Event Format
Online only: The session will take place exclusively via Microsoft Teams.
The access link will be shared with all registered participants before the event.
🎒 What You Need
An interest in modern data platforms, analytics engineering, or BI
Basic knowledge of dbt, Trino, or BI tools is helpful but not required
A laptop or desktop computer with internet access
A Microsoft account (recommended)
🎤 About the Speaker
Egor Tarasenko is a Data & AI Engineer at Ponder with more than five years of industry experience.
At Ponder, he helps deliver data and AI services tailored to educational institutions, with a particular focus on building practical solutions that make data more accessible and valuable.
Egor is also actively involved in the open-source community, contributing to AI and data libraries as well as plugins and integrations for technologies such as Apache Superset, Apache Airflow, dbt, Trino, and Apache Iceberg.
🙌 Join Us
Join the Data Monsters Meetup for a practical session with Egor Tarasenko and discover how Trino and dbt can help prevent upstream changes from unexpectedly breaking critical dashboards.
Whether you're working with Superset, Tableau, Power BI, or another BI platform, you'll leave with a practical approach for making dependencies visible, validating changes earlier, and building more reliable analytics workflows.