TAFI, AI-assisted building
AI-assisted building. Your application model. Your approval.
TAFI is our AI assistant in TAF Studio. It works on the same application model you inspect and extend, proposes a plan, and applies nothing until you confirm. It does not replace architecture decisions, code review or your team's responsibility for production readiness.
- Clarifying questionsWho approves leave? Do half days count?WorkingAnswered
- RequirementsLeave requests, balances, manager approvalWorkingApproved by you
- BlueprintEntities, screens, lifecycle and workflowsWorkingApproved by you
- Build planEvery change TAFI will make, in orderWaiting for you
- Live build and scoreStarts only after you approve the planNot started
How It Works
Confirmation is enforced by the platform, not left to the AI to remember.
Product Capabilities
Whatever TAFI builds is ordinary TechAppForce configuration your team can read, change and release.
The same app model
More
Entities, queries, screens, lifecycles, workflows, permissions and menus, stored as if a developer had built them by hand.Reviewed in TAF Studio
More
Open anything TAFI created in its designer. Read it, change it or delete it.Released like any change
More
TAFI builds in development. Its work reaches customers only through a release you review.Checked before it reports
More
Each build step is checked before TAFI moves on. A step that reports an object it never created is rejected.
Frequently Asked Questions
Where can I use TAFI?
Built into TAF Studio, and available to Claude Code, Cursor or Codex through a gateway that speaks MCP. The same checks apply everywhere.
Who is TAFI acting as?
The person who signed in, and only what that person's role allows. Every change is recorded under their name.
What if a rule is too specific for configuration?
Your developers add a C# script as a record hook or workflow action, or a TypeScript screen extension, in the same app.
Is anything saved without my approval?
No. Interactive changes run as a dry run first, and nothing is saved until you confirm.
See TechAppForce on your own use case
Bring a real requirement. We run TAFI on it with you, from the first question to live screens in a development environment.