AI Coding & App Building for SMBs · Build with AI
Human Review, Testing, and AI QA for SMB-Built Apps
Apply the five-step QA framework for AI-built apps - from acceptance criteria to auth testing, rollback discipline, and handoff documentation.
8 lessons · 120 minutes
Before starting: A deployed or nearly deployed app built with an AI app builder or IDE coding assistant · Completed Build Internal Tools with Lovable and Base44 or equivalent hands-on building
Start courseWhat you walk away with
- Write acceptance criteria in plain business language before testing starts
- Test an AI-built app with real data: broken inputs, edge cases, and wrong user roles
- Verify auth, permissions, and data isolation checks on anything that touches customer records
- Undo a bad deploy with rollback discipline steps that preserve data
- Produce handoff documentation and a maintenance plan, including the rebuild or migrate decision
- A build-ready app spec you can export to Lovable, Base44, Cursor, or Claude Code. Every artifact saves to your workspace, so you keep it after the course ends.
- A knowledge-check quiz and a visual walkthrough inside every one of the 8 lessons, about 2 hours of guided work.
- Most lessons close with a checklist, template, or do-it-this-week task, so progress shows up in your work, not just in the app.
Why AI-Built Apps Fail in Production: The 60-70% Problem
Understand the research-backed reality of where AI app builders reliably succeed and where they fail - with five concrete production failure scenarios - so you can build with accurate expectations and avoid the most costly mistakes.
Writing Acceptance Criteria in Plain Business Language
Learn to write acceptance criteria in plain English using the "When / the system should / and" format - the most teachable QA skill that gives you a testable checklist for every version of your app.
Testing with Real Data: Broken Inputs, Edge Cases, Wrong User Roles
Execute your acceptance criteria checklist with systematic test scenarios - broken inputs, edge cases, wrong user roles, and mobile - and document every failure as a fix item before any customer sees the app.
Auth, Permissions, and Data Isolation: The Non-Negotiable Checks
Run the four critical auth and permissions checks that every deployed app needs - can users see each other's data, can non-admins take admin actions, can unauthenticated requests reach protected data - before any customer-facing deployment.
Rollback Discipline: How to Undo a Bad Deploy Without Losing Data
Build the rollback habits - screenshot, export, version control, and communication plan - that let you recover quickly from a bad deployment without losing data or leaving your team in the dark.
Documenting What You Built (So You Can Hand It to a Contractor)
Transform your Build Brief into a complete app documentation package - what is stored, what is connected, what it does, and what was intentionally left out - so any developer can pick up your project and continue without starting over.
Maintenance: Updating and Iterating Your Tool as the Business Changes
Establish the operational habits that keep your internal tool working reliably over time - prompt-based updates, the test-after-every-change rule, and documentation maintenance as the app evolves.
When to Rebuild vs. When to Migrate to a Proper Platform
Apply concrete exit criteria - user count, data sensitivity, reliability requirements, customer-facing status - to decide when your AI-built tool has outgrown its platform and what the migration path looks like.