Ask HN: Where do you store SQL and does it cut down "can you pull this?" pings?
1. *Saving & sharing* – queries end up everywhere (Slack, Git repos, BI folders, Notion pages…).
2. *Endless requests* – because nobody knows what exists, engineers and data folks still get hammered with “please run this query / build this dashboard” tickets.
How is your team handling both sides?
- Do you have a single, searchable library? - Can non-devs discover and trust queries without bugging you? - What’s failed, what’s stuck?
---
We got frustrated enough to build *Galaxy*:
- *Cursor-style editor* → fast, modern, context-aware AI copilot for querying. - *Collections* → shared, version-controlled query library (like Postman). - *Endorsements* → mark a query as “Endorsed” so the next person stops asking.
Tiny pre-alpha opens tonight — brutal feedback welcome. We are excited to reimagine the SQL editor from the ground up and bring Datagrip, Dbeaver, Tableplus, etc into the modern age.
- Reddit threads that sparked this: - https://www.reddit.com/r/SQL/comments/1kzdn2p/does_your_team_have_a_sql_library_or_just_chaos/ - https://www.reddit.com/r/dataengineering/comments/1l049w2/how_do_you_push_back_on_endless_urgent_data/
• Feature overview: https://getgalaxy.io/features/collaboration
Would love to hear your war stories (and see if our approach actually helps).
I think the most disciplined organizations use stored procedures but the ordinary methods people use for managing those often date from before version control. (e.g. so many people would wrongly laugh at stored proc users for being so 1990s + Oracle)