The Problem: Smart prototypes often fail silently due to unmonitored thermal spikes or loose relay contacts causing unexpected downtime.
The Fix: I engineered a microcontroller watchdog script paired with sensor feedback to track environmental thresholds and automatically trigger physical buzzer alerts before failures occur. ⚡
What sells it for me is the refusal telling the agent exactly where the token should go. A bare "no hex" error usually just sends it looping. Does guard.mjs run only as an agent hook, or in pre-commit too so it catches humans as well?
Lately I get a lot of requests for product audits (in terms of UX/UI and sometimes even with suggestions for CX!!). Most of them are new products built fast, often with AI
I just finished 1 for a train booking site: 2 reports, mobile and desktop, 54 findings. Every finding has a number, a screenshot with the problem marked, why it matters to the customer, the exact fix and a priority
The format is the point - > you can hand it to your developer or straight to Claude and it can apply the fixes one by one
If you're about to launch something new and want a designer's eyes on it first, send me a message 😉
Btw, how often do you audit your product? I think it should happen before every launch and in the process with received data