← sources

Building a SaaS to navigate Schema changes and prevent broken dashboards. Work in progress. Looking for feedback.

source: reddit (reddit:SaaS)published 2026-10-23 01:50ingested 2026-10-24 02:37

retained text

I’m a data engineer, and I've been building a SaaS called VraieSource, which helps with identifying downstream impacts on data workloads end-to-end through the transformation logic to the BI reports from an upstream change in source tables, such as column drops, renames, etc.

In a small data stack, the migration might be in GitHub, the transformations in dbt, and the reports in Metabase. Tracing a column through all three by hand takes time, especially when you need to know which specific field in a report is affected.

I’m building VraieSource to run that impact check on a GitHub PR. It would read the proposed schema change, follow column dependencies through the transformation SQL, and show the affected report fields, along with anything it couldn’t verify.

I’ve made an interactive demo so people can try that analysis before the live integrations are ready. It uses sample PostgreSQL, dbt, and report inputs; it doesn’t connect to your GitHub repo or data tools yet.

Could a few of you try the rename or drop scenario and tell me whether the result would help you review a real change? I’d especially like to know what evidence you’d need before trusting a check in a PR. I'm trying to gauge if this is a real problem that people are looking to solve basically :P

Demo: [https://vraiesource.lovable.app/workflow-demo/](https://vraiesource.lovable.app/workflow-demo/) (this is just a prototype)

This is my first time building a product, so I’d also welcome advice from anyone who’s brought a developer tool to its first users.

Thanks!

document fetched 2026-10-24 00:29 · canonical https://reddit.com/r/SaaS/comments/1xd16zh/building_a_saas_to_navigate_schema_changes_and