This Portfolio —
Built with Claude

Designing and building my portfolio through conversation with AI.

5
Case studies shaped with Claude
~0
Lines of code written by hand
100%
Built through structured prompting, iteration, and review

A portfolio about using AI, built using AI

This portfolio is not just a place to talk about how I use AI in my design process — it is also an example of that process in practice.

The site you are reading now was designed and built entirely through conversation with Claude. There was no traditional design handoff, no separate development workflow, and no manual implementation from scratch. Instead, the process was structured, conversational, and iterative.

I began with a clear brief, defining the content, tone, and direction of the site. From there, I worked directly with Claude to translate that into a functioning product — reviewing outputs, refining details, and iterating in real time.

This way of working mirrors how I already approach product design. AI is not a replacement for design thinking, but a tool for accelerating execution — enabling faster iteration, broader exploration, and a shorter path from idea to working product.

The judgement, prioritisation, and design intent remained mine throughout. Claude acted as a flexible implementation partner.

In that sense, this portfolio is both a finished product and a working example of how I design with AI.

Structured, conversational, and iterative

The process started with a well-defined brief. Before building anything, I created a written document outlining the structure, content, and overall direction of the portfolio.

Claude was then used to translate that brief into a working site. Once a foundation was in place, the process became highly iterative — refining layouts, rewriting sections, and improving interaction details through smaller, targeted prompts.

Working this way felt closer to collaborating with a developer than using a traditional design tool. I would describe an intention, Claude would implement it, and I would review the result in a live environment — either approving it or refining the direction.

Some decisions were resolved quickly. Others required multiple rounds of iteration to reach the right outcome. That back-and-forth was not separate from the design process — it was the design process.

Using AI within real design workflows

While this portfolio was built as a standalone project, the same approach has been applied within product design work — particularly in early-stage exploration and iteration.

In one case study, I used AI alongside the design process to:

Explore multiple interaction patterns before committing to a direction

Generate and refine content for complex user flows

Prototype and test ideas more quickly than a traditional design–development cycle

This made it possible to move from initial concept to a testable solution faster, while maintaining control over the overall experience and design decisions.

Rather than replacing the design process, AI became a layer within it — supporting exploration, accelerating execution, and making iteration more fluid.

What this way of working revealed

The most valuable aspect of this process was not just speed, but clarity.

Working through conversation required precision. When prompts were vague, the output reflected that. When direction was clear, results improved quickly. The process became a way of refining both the design and the thinking behind it.

It also shifted where time was spent. Less time went into manual production, and more into structure, content, and decision-making — which are often the most critical parts of design.

This portfolio took significantly less time to build than a traditional workflow, but it did not feel less considered. If anything, it made the decision-making process more explicit.

How I approach design more broadly

Using AI to accelerate execution — moving from brief to working product quickly without compromising quality.

Staying hands-on in interaction and detail — reviewing outputs closely and refining where needed.

Iterating deliberately — treating each round of feedback as part of the design process.

Keeping design judgement at the centre — AI supports execution, but decisions about what to build and why remain human.