# Non-dev building a multi-tenant clinic SaaS with AI agents. What should I learn first? — retained record

> Retained reddit record from reddit:SaaS.

Canonical: https://highsignal.app/data/records/f4ffd5bc5ddce646

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:SaaS)  published 2026-11-19 02:46 ingested 2026-11-20 02:37

[original source ↗](https://www.reddit.com/r/SaaS/comments/1xzyfgj/nondev_building_a_multitenant_clinic_saas_with_ai/)

## retained text



```
Hi! I manage a medical clinic and I’m building a multi-tenant SaaS now (appointments, patient records, invoicing, etc.) using AI coding agents (Claude Code). Planned stack: React + TypeScript + Node + PostgreSQL.

My background is not in tech, I don’t want to write the code myself, but I want to understand enough to get the backend right (data, security, tenant isolation) and not mess things up early.

**A few questions:**

1. What are the best in-depth resources (paid courses included) for database design and multi-tenant architecture?
2. What backend mistakes are hardest to fix later in a SaaS like this?
3. How do you run multiple AI agents in parallel without them breaking each other’s work? (roles, worktrees, tickets, reviews, CI)
4. How do you keep agents from drifting, e.g. ignoring the architecture or touching sensitive code (auth, tenant isolation)?

Also, if anyone experienced would be up for explaining things 1:1 from time to time (paid), feel free to DM me.

Thanks!
```



document fetched 2026-11-20 00:28 · canonical https://reddit.com/r/SaaS/comments/1xzyfgj/nondev_building_a_multitenant_clinic_saas_with_ai
