I'm Radim Marek. I speak internationally on PostgreSQL internals, write boringSQL, and help teams fix production database problems. Twenty years ago I was a Hibernate trainer, helping developers keep databases at arm's length. Databases were a necessary evil, a dumb storage layer you tolerated, not something you learned on purpose.
Then I actually learned SQL. And PostgreSQL. And everything I thought I knew about where to put logic, how to model data, what "scales" really means, turned out to be wrong. Not slightly off. Fundamentally wrong.
Turns out the technology I was running away from is the one part of the stack that survived every framework migration, every rewrite, every paradigm shift. Java to Rails to Python to Go. REST to GraphQL. Monoliths to microservices and back. The database outlasted all of it.
boringSQL is the site I wish existed when I started taking databases seriously. Practical, hands-on, no fluff. The kind of content that helps you understand what actually happens when you DELETE a row, why VACUUM lies to you about your indexes, or what lives inside an 8KB page.
I build tools to solve production problems I've hit myself. DryRun gives AI agents your real PostgreSQL schema without handing over production credentials. It captures schema, stats, and query workload into a file, then runs migration safety checks and schema linting offline, in CI, where the dangerous ALTER gets caught in review instead of the incident channel.
Alongside it - SQL Labs for safe hands-on practice, RegreSQL to catch query regressions before they ship, and Fixturize for test database snapshots that don't slow your CI to a crawl.
The training habit never went away, it just got a better subject. Workshops, an annual training programme and PostgreSQL consulting run through Clusterity, the company behind boringSQL and the tools above. If your team ships migrations faster than anyone can review them, that is what those are for.
If any of this sounds useful, subscribe to the newsletter or drop me an email at radim@boringsql.com.