One specific internal tool, hardcoded to your own database, is a genuine weekend build. A general-purpose drag-and-drop builder that works for anyone's data is a much bigger, more open-ended product.
Borderline
Effort: a weekendRunning cost: $0/mo to start
Why CanICodeThis says Borderline
Connecting a data table view and a form directly to your own database queries is a well-documented admin-panel pattern
For one fixed internal tool (not a general builder), hardcoding the layout and queries is genuinely a weekend
Basic role-based login for an internal tool is standard auth
The parts that'll cause problems
A general-purpose visual builder that works for any database schema and any layout is a much bigger, open-ended engineering problem than one fixed tool
Connecting to many different data source types (not just your own database) multiplies the integration work
Version control and safe editing by non-engineers is real platform depth beyond a single hardcoded tool
How I'd build this
frontend
Next.js
database
Supabase
auth
Supabase Auth
hosting
Vercel
MVP scope
One specific internal tool: a data table with filters and an edit form
Connected directly to your own database
Basic role-based login
Postpone to v2
A general drag-and-drop builder for arbitrary future tools
Multiple data source connectors
Non-engineer-safe editing with version history
What it actually costs to build
AI agent — with a subscription (Claude Pro/Max, Cursor, etc.)
$0 marginal
AI agent — pay-per-use API, no subscription
$43–$86 one-time
Hosting, once it's running
$0/mo (free tier)
Domain name, if you want your own
~$12/yr
Existing tools solving a similar problem typically run $10–$24/mo — see how they compare below.
The build prompt
Don't build a general internal-tool platform. Build one specific internal tool instead: an admin page with a data table from your own database, filters, and a form to edit or create records, with basic role-based login. That covers a real, single internal need directly.