AI in practice
Role: Senior Product Designer
Skills: AI, Product design, UX, Prototyping, Front-end experimentation
Company: Independent / Self-initiated
Over the last year I've been working hands-on with AI across design and code. Pairing tools like ChatGPT, Cursor, OpenAI and Figma Make, I've been able to move from idea to working experiments much faster.
Rather than stopping at concepts, I've been building and testing ideas in code. This has helped validate product decisions earlier and close the gap between UX thinking and real implementation.
The focus has been on practical use. Small experiments that explore how AI can support everyday product experiences, from structured inputs to more adaptive, assistive flows.
Designing and shipping with AI
The shift has been from static design outputs to working product experiments. Using AI as part of the workflow allows faster iteration, earlier validation, and more informed design decisions.
It enables testing not just how something looks, but how it behaves.
- Frame the problem clearly
- Explore directions quickly using AI
- Prototype interactions and flows
- Build working experiments in code
- Test, learn and refine
- ChatGPT for thinking and iteration
- Cursor for building features
- OpenAI APIs for behaviour and logic
- Figma Make for rapid UI exploration
Building an AI-assisted grocery planning web application, including ingredient parsing, smart suggestions, and lightweight forecasting.
- Built working list creation and parsing flows
- Tested AI-driven suggestions based on context
- Explored planning and budgeting concepts
I used Figma Make to rapidly prototype and pressure-test the scope for an AI-assisted grocery app. The concept covers end-to-end moments: creating lists, generating recipe matches, scanning ingredients while shopping, and finding nearby local suppliers.
The core list was intentionally kept lightweight: quick entry, clear scanning, and thumb-friendly interactions for one-handed use. A subtle completion celebration adds delight at the end of the journey.
To reduce the friction of recipe-to-list planning, I shipped a recipe URL import feature using Cursor + OpenAI. It parses the recipe into individual ingredients and converts measures into metric units such as milliliters or grams, making it faster to shop accurately.
Investigating how AI can support funding discovery and comparison for small businesses.
- AI-assisted search and filtering
- Structured comparison outputs
- Visual exploration of funding options
Early exploration of adding an AI layer to help explain work and provide additional context.
- Structured content prompts
- Context-aware responses
- Lightweight integration approach
- Rapid prototyping beyond static design
- Translating ideas into working features
- Reducing risk through early validation
- Strong alignment between UX and implementation