~/ai-playground
Product Designer · AI-Assisted Product Development
Exploring the intersection of product thinking, systems design, and AI-enabled building.
01 / Intro
Over the past year, AI has gradually become part of how I think through product problems, explore ideas, and build faster.
Most of these projects started as experiments. Sometimes I was trying to solve a real operational problem. Other times I was exploring a workflow, testing an idea, or trying to understand how AI changes the process of designing and building products.
What interests me most is not just generating outputs — but using AI to reduce friction between idea, exploration, and execution.
02 / Method
A repeatable system for idea to execution.
01 / Research
Gathering signals. Organizing research. Making sense of domain knowledge before generating solutions.
02 / Exploration
Pressure-testing ideas. Asking sharp questions. Surfacing the edges of a problem space.
03 / Prototyping
Making ideas tangible. Fast, rough, real. Optimizing for learning speed over polish.
04 / Refinement
Introducing structure. Resolving system logic. Turning rough surfaces into considered ones.
05 / Iteration
Feedback loops. Repeated cycles. Each pass makes the thinking sharper.
Tools in rotation
03 / Work
Built to solve real problems. Designed to understand what AI changes about the process.
Why I built it
To replace fragmented manual tracking systems and explore how quickly operational software could be prototyped with AI-assisted workflows.
What I explored
Structured data modeling. Internal tool UX patterns. The speed ceiling of AI-assisted prototyping.
Process
04 / Writing & Talks
Thinking in public.
How to design AI products that are useful, trustworthy, and human-aware without overpromising or misleading users.
A beginner-friendly walkthrough of vibe-coding an app using Figma Make, packed with practical tips and clear steps.
A recorded talk from the Veet Career Series 2025 where I shared practical insights on staying relevant, adaptable, and valuable in an AI-driven tech landscape.
05 / Notes
Thinking out loud.
2026.03
The most interesting thing about AI-assisted building isn't the speed. It's how quickly you can make something real enough to react to. That feedback loop changes how you think about ideas.
2026.02
Every AI workflow still breaks down without clear thinking at the start. Garbage in, garbage out — but faster. The craft is in the prompting, the structure, and knowing when to stop.
2026.01
Most conversational products feel like they were designed by engineers. There's a lot of whitespace in designing for voice and text interfaces that actually feel considered.
2025.12
Some of the most-used software in the world is the least designed. Internal tools, tracking systems, operational dashboards — there's a huge opportunity in that space.