Literally no reason to use Supabase in 2026. I will not promote.

Hi, Just wanted to share my own personal experience as a senior full stack developer. Loads of experience across different stacks and architectures and here are my concerns with Supabase, having actually rewritten a commercial project to use it. Regretted it hard. Supabase is a weak choice to build a clean codebase with proper separation of concerns and CI/CD. Especially in this day and age with AI, there just aren't any reasons why one would use RLS and Edge functions to define critical business logic and not own the auth schema. An after insert trigger on auth.users and surfacing every failure as a vague 500? Never again am I buying into these awful anti-patterns. Personally, found even a stack like PHP Symfony with Doctrine to be a way better developer experience in 2026, although my go-to these days is a monorepo using nestjs, ts-rest, nextjs and react native. Expect a lot of hate for this, but Supabase is a massive marketing charade and SQL was never meant to go this far. PostgREST is an interesting experiment but c'mon any serious company will use Spring Boot or .NET. Literally anything but this. There is no gain in velocity, you're building a house on weaker grounds for absolutely no reason other than a lack of experience and/or understanding the consequences. I will get downvoted by supabase fanboys and people that don't understand what I'm saying and that's okay, but if I convince at least one founder not to use it, I've done my job. Thanks for reading and happy coding.