Vibe Coding Rescue

Vibe-Coded App Backups: You Don’t Have One Until You’ve Restored It

7 min read | Updated September 2026

“Do you have backups?” is a question with a trick in it. Almost everyone answers yes — the hosting provider does them, the database has a snapshot feature, the AI said it set something up. The real question is: when did you last restore one, and how long did it take? If the answer is “never,” you don’t have a backup. You have a hope, and a vibe-coded app is where hopes get tested — usually by a migration the AI ran on production.

The two scenarios you’re actually preparing for

The AI dropped a table. A confident “I’ll clean up the unused columns” that wasn’t unused. You need to get back to five minutes ago, for one table, without losing the last hour of everyone else’s data. The account got compromised. Someone deleted or encrypted what they could reach. You need a copy that lives somewhere they couldn’t reach, from before they got in.

What “a backup” has to mean

1
It exists automatically, on a schedule you didn’t have to remember
Daily at minimum for any app with paying users. Check what your database tier actually includes — free and hobby tiers often include nothing, or only a rolling few days.
2
It can be restored to a point in time, not just “last night”
A nightly snapshot loses today. Point-in-time recovery (PITR) — continuous backup of the transaction log — lets you go back to 14:07, right before the bad migration. On Supabase and most managed Postgres hosts it’s a paid add-on worth every cent once money flows through the app.
3
At least one copy is somewhere else
A backup in the same account as the database dies with the account. Export a periodic dump to a separate cloud storage bucket under separate credentials — a weekly pg_dump to S3/R2/GCS is enough for most.
4
It covers the files, not just the database
User uploads in Storage / S3 / Firebase Storage are not in the database backup. Turn on bucket versioning or replicate to a second bucket.
5
It covers the config
Environment variables, DNS, the list of third-party services and who has access — the things that make the restored database into a running app again. Keep them in a password manager or an encrypted note, not in one person’s head. The handoff checklist is a good template for what to capture.
6
You have restored it, timed it, and written the steps down
On a staging project, from the real backup. Note how long it took. If it took three hours of figuring out, do it again next month until it’s thirty minutes of following a list.

The restore drill (do this once a quarter)

Create a fresh staging project. Restore last night’s backup into it. Point a staging deployment at it. Log in, look at real (or anonymised) data, run the five core tests. Time the whole thing. Then delete the staging project. The first time this takes an afternoon and reveals two things the backup didn’t include. That afternoon is the cheapest insurance you’ll ever buy.

Prevention: don’t let the AI touch production data unsupervised

Most restores we’ve done for founders weren’t after attacks — they were after an AI tool ran a migration, a “cleanup,” or a seed script against the live database. Three rules that stop it: production credentials never in the local .env the AI can see; every schema change goes through a migration file that runs on staging first; and “never run migrations, deletes or bulk updates without asking” written into the tool’s instructions. The schema-smells guide covers how to sequence the migrations you do want.

What to do right now if something’s already gone

Stop writing to the database if you can (maintenance mode). Don’t run anything “to fix it” — every write makes the point-in-time target harder to hit. Note the exact time it happened. Check whether your database has PITR or a snapshot from before that time. If yes, restore to a new instance, verify the missing data is there, then decide how to merge it with what happened since — that last step is where an engineer earns their day rate, and it’s worth getting one on a call before you overwrite anything. If no backup exists: check whether the provider keeps anything you can’t see (support tickets sometimes recover more than the dashboard offers), and check whether any other system — email receipts, payment provider, analytics — holds a partial copy.

FAQ

My host says it backs up daily. Isn’t that enough?
It’s the minimum. Ask three questions: how many days are kept, can I restore to a point in time, and can I download a copy. If any answer is “no” or “not on your plan,” add an external dump.
Does Firebase / Firestore back itself up?
Firestore offers scheduled backups and PITR as features you enable — they are not on by default. Realtime Database has automated backups on paid plans. Turn them on and export to a bucket you control.
How much does proper backup cost?
Usually a small fraction of your database bill for PITR plus pennies for external storage. The restore drill costs an afternoon. Compare with the cost of rebuilding customer data from memory.
Can the AI set all this up?
The export script and the bucket, yes. Enabling PITR is a dashboard setting. The restore drill is the part only a human can do — because the point is to find out what the AI’s setup missed.

Want the restore drill done with you?

Half a day: we confirm what your backups actually contain, enable point-in-time recovery, add an off-site export, and run a timed restore into staging with you watching.

Book a Backup Drill →

Engineering Insights

Latest from Syntaxa Studio.

Loading latest posts