Director of Product Design

EDWARD
KARCHER

Edward Karcher
Phoenix, AZ  ·  ekarcher@gmail.com  ·  480.677.0272
Inflection Points →

I build AI products in production, not prototypes of them. My current focus is Presence at GoDaddy, spanning AI-native and manual website builders, WordPress, and Hosting, roughly $1.25B in annual revenue.

AI amplifies whatever intent you give it. Vague context does not stay contained, it compounds until it erodes the system and destroys customer confidence. That is why the work I find most interesting lives at the system level: the shared language that keeps AI output consistent, the constraints embedded before execution starts, the operating models that hold without someone watching. That shapes how I lead, how I build teams, and how I think about what sits underneath the products.

Recent Impact
+60%
QoQ Installs
New mobile product launch
+20%
Total GCR growth
Revenue growth
+600%
Revenue growth
+$250M, 360K customers migrated
+$700k
Incremental GCR
Via UX-led initiative
+21%
Inline Editing Engagement
Background Color Controls
+10%
Conversion
Airo for WordPress

HOW I THINK ABOUT THIS WORK

AI does not behave the way software used to behave. It drifts. It infers. It produces confident output from ambiguous input, and that changes what leadership is responsible for. The failure mode is not moving too slowly. It is moving fast with weak intent and finding out in production. Small gaps in context compound. A confident tone from a system that does not actually know what it is doing erodes trust faster than any bug.


My focus is building the conditions where that does not happen. Explicit intent. Constraints embedded before execution starts. Feedback loops that catch drift early enough to correct it. That is what I spend my time on.

01
SYSTEMS

Quality does not hold because someone is watching. It holds because the constraints are already there. I build standards into the system itself, in the language model, in the component library, in the way work gets handed off, so quality is a property of the infrastructure, not a function of proximity.

02
DECISIONS

I do not wait for certainty before shipping. I define the decision clearly, make it reversible where possible, and treat production as the best feedback loop available. Behavior in the wild tells you things pre-release testing cannot. The faster you get signal, the faster you correct.

03
TEAMS

Teams move well when expectations are explicit and ownership is clear. I spend more time on those two things than anything else. Approval chains shrink when boundaries are understood. Accountability does not require surveillance when people know what they are responsible for and why it matters.

Without discipline, speed stops being an advantage and becomes a liability.

HOW THE WORK MOVES

I start with intent. I define the outcome and resist the urge to prescribe how we get there, because clarity of direction matters more than control of execution. I embed constraints early so they protect quality without requiring oversight. I ship in order to learn, because production exposes assumptions faster and more honestly than any planning cycle.

The work holds longer when the discipline is built in, not bolted on.

This site was built the same way I build products. Every page, every line of copy, every image produced by AI agents directed by a single human in the loop. The process is documented here.

Connect with me
480.677.0272
ekarcher@gmail.com

The work I have been doing at GoDaddy is not a detour. Building AI systems in production, teaching models customer vocabulary, shipping directly in code. That is exactly the kind of work I want to keep doing, at the companies building the infrastructure underneath it.