Kustomer
/Head of Product Design
When engineering stopped waiting for design
Engineering had embraced AI and was shipping faster than ever. But they were shipping without design — pulling components off an outdated design system, making UX decisions in code, and routing around a team they saw as a bottleneck. The product was moving fast. The experience was getting worse.
Context
The market pressure to adopt AI was intensifying. Competitors were claiming 10–20x velocity improvements. The CEO was pushing more code than anyone in the organization. Engineering had AI in their workflow. Design didn’t — and was being excluded from decisions that shaped the product.
The design system was underfunded and outdated. Engineering pulled patterns off the shelf that weren’t always appropriate, creating rework for both teams. Reporting — the #1 reason for lost deals — had been stuck in the backlog for months. The longer design was out of the loop, the harder it was to get back in.
Approach
Proved it on the highest-stakes project. Took reporting — the most-requested feature and primary deal-loss driver — and partnered with the most senior designer to rebuild it in weeks instead of quarters, without engineering support. Used an AI-accelerated workflow (Claude for design logic, GitHub for implementation, Storybook for validation) to demonstrate that design could ship at engineering speed on its own.
Showed the work. Documented the process publicly — what worked, what didn’t, how the new workflow changed the output. This wasn’t a mandate to adopt new tools. It was evidence that design could move at engineering speed without sacrificing judgment. The designers who saw the results pulled themselves in.
Restructured how design plugged into the pipeline. Defined how PM-first and engineering-first prototypes would flow through design for quality review — so design stopped being a gate and started being a collaborator at the speed the org needed.
Tradeoffs
Timing
Should have adopted earlier. Nine months before, at an offsite, the team had recognized AI was entering the space and built early designs. But we defaulted back to Figma workflows to keep up with the roadmap. The urgency of the present kept delaying the investment in the future.
Design system debt
Should have pushed harder on getting the design system funded. An up-to-date system is fuel for AI design workflows — without it, AI generates plausible but wrong patterns. The underfunded system created rework that a maintained one would have prevented.
Adoption gap
Not every designer adopted at the same pace. The team members who leaned in got dramatically faster. Those who didn’t fell further behind. Managing that gap required more coaching than anticipated.
Outcomes
Resolved
#1 deal-loss driver
Reporting — the primary reason for lost deals — redesigned and shipped in weeks instead of quarters
Bottleneck → Driver
Design perception
Engineering went from routing around design to pulling design in early
“He created an environment where design didn’t just complement engineering — it truly pushed the entire product forward.”
Michael Blake, VP of Engineering, Kustomer