tl;dr

Highly recommend. Using AI tools like large language models and AI-assisted coding environments, I was able to accelerate requirement writing by drafting initial descriptions for each feature and then asking for suggestions on what else to include. It even helped me revisit some dusty front-end coding skills I hadn’t used in years.

Context

I’m in a hybrid product management and design role at an AI-centric HealthTech startup. I’ve been here for about 5.5 years.

During that time, I helped design and build the platform’s first user interfaces, turning a lot of manual processes into scalable, user-friendly tools.

Before this role, I was typically hands-on with both designing and coding front-ends — a practice that shaped how I now approach product development in more collaborative, cross-functional settings. These days, I collaborate more closely with engineers, behavioral scientists, clients, and company leadership to create intuitive, omnichannel experiences that actually move the needle on patient engagement and outcomes.

That said, my web dev skills have definitely gotten a bit rusty from lack of use — which made this project especially fun.

The project

An internal tool to help our operations team assess whether our interventions are ready to launch — who they’ll reach, what’s being sent, and whether anything’s blocking the way.

It empowers non-technical team members to independently monitor readiness, investigate potential issues, and take action before anything impacts patients, members, or consumers.

It brings clarity and speed to a critical step in our process — helping ensure every intervention we launch is fully aligned and operationally sound.

My role

I designed the experience from the ground up in close collaboration with key stakeholders, including the operations team leader, to make sure we were solving the right problems in the right ways.

I facilitated cross-functional brainstorming sessions with both technical and non-technical team members to align on goals, surface edge cases, and shape the direction together.

While I followed a traditional design process and leveraged short, highly iterative design-to-feedback cycles, I also used AI tools like Cursor to help me ramp back up on Angular and prototype front-end components that aren’t yet part of our existing UI library. Seeing the designs come to life in a working prototype helped me quickly identify areas for improvement. Those insights led to revisions in both the design and its underlying requirements.

One of the most surprising and valuable uses of AI, particularly large language models, came during the requirement writing phase. I wrote a few paragraphs describing each feature and used AI to suggest additional details and edge cases I might not have considered. I was genuinely impressed by the quality of the ideas — it felt like having a second brain during early-stage thinking. I also used AI to help generate acceptance criteria for each feature, based on the requirements we had outlined together, which made handoff and planning even smoother.

The tool is still in development, but the design work already serves as a foundation for future client-facing features that will bring similar visibility and control to our partners.

Final thought

This project was a glimpse into a new way of working — one that blends traditional design methods with the speed and support of modern tools like AI.

It made the work faster, more collaborative, and surprisingly enjoyable.

If you’re in a product or design role, now’s a great time to start exploring these kinds of approaches. I’d love to hear how others are integrating AI or working differently — share your experiences if you’re on a similar path.