One workflow I’ve been experimenting with is using AI throughout the journey from UX design to a working application.
1. Understand the requirement
I start by giving AI the business requirement, user roles and process documentation. I use it to break down the workflow and identify missing scenarios before designing.
2. UX exploration
I use AI to explore user flows, information architecture, edge cases and different interaction approaches. This helps me think beyond just the happy path.
3. Design in Figma
Once the flow is clear, I create the UI in Figma and use AI to review the screens for consistency, usability and possible edge cases.
4. Design → Code
After the design is finalized, I use AI coding tools to help translate the design into frontend components. Instead of blindly generating everything, I review the structure and make sure it follows the project’s existing patterns.
5. Test & refine
I use AI again to help identify possible UI issues, missing states, validation cases and responsive behaviour.
6. Build & deploy
Once the implementation is ready, the application goes through the normal build and deployment process. AI can also help explain build errors, configuration issues and deployment problems when they come up.
What I like about this approach is that AI isn’t being used for just one step. It acts more like a supporting layer across the whole journey:
Requirement → UX → Design → Development → Testing → Deployment
The important part for me is still keeping human review at each stage rather than treating the AI output as the final answer.