Designers Who Vibe Code Are Happier At Work
TL;DR
Design is splitting into three camps — Tommy Geoco’s March survey of 1,478 designers found 43.8% are heavy AI coders, 18.5% are dabblers, and 37% are complete holdouts, creating what he calls “three parallel realities” inside the profession.
AI tools are now core to designer workflows, not side experiments — In the top 10 weekly tools, Claude ranks just behind Figma at #2 and Claude Code sits at #4 ahead of FigJam, while 71% of respondents say they’ve either added AI or gone “AI central” in the last six months.
Designers who vibe code report being meaningfully happier at work — Non-AI coders rated job satisfaction at 5.93/10 versus 7.39/10 for heavy vibe coders, a gap Geoco says surprised him given how negative online discourse around AI can be.
Managers are adopting AI faster than many individual contributors — About 46.6% of managers/directors report vibe coding compared with roughly 35% of traditional IC designers, which leads Geoco to argue that managers who don’t make experimentation time for their teams become the bottleneck.
Designers aren’t just using AI tools; they’re building their own — 59% say they built a tool, app, or utility in the last six months, one in four does it regularly, and Geoco highlights examples from a non-coder making a Premiere Pro plugin with Claude Code to Ramp shipping 1,500 internal apps in six weeks.
The biggest blockers are practical, not ideological — The top reasons designers aren’t adopting AI coding are lack of time (57%), tool overload (53%), and concerns about output quality (52%), with many teams still stuck in “lunch and learn” mode rather than real workflow change.
The Breakdown
A survey big enough that the DOJ called
Geoco opens by grounding the data: UX Tools has run annual design tool surveys since 2017, drawing from a broad base of nearly 100,000 designers across startups, enterprise, healthcare, banking, government, and 18 regions worldwide. He reminds you this isn’t just SF-and-NYC tech hype, noting the surreal flex that the U.S. Department of Justice used their data during the Adobe-Figma audit.
The tool stack changed fast — and Claude is now everywhere
This quarter’s prototyping survey shows five of the top 10 weekly tools are now AI-first. The standout stat is Claude at #2 right behind Figma, with Claude Code at #4 ahead of FigJam — which Geoco points out had dominated the whiteboarding category just last year, making the speed of the shift feel almost absurd.
Heavy AI coders, dabblers, and holdouts are living in different worlds
The headline split: 43.8% of respondents use AI coding tools for more than half their building time, 18.5% are dabbling, and 37% haven’t touched vibe coding at all. Geoco’s point is that this isn’t a simple early-adopter curve — it’s “three parallel realities,” and many designers aren’t even sure which one they’re standing in.
Who’s adopting fastest, and why that changes the design role
Design engineers lead adoption at roughly 80%, followed by lead/principal designers at about 56%, while traditional IC product designers sit closer to 35% and researchers trail at around 26%. He pairs that with Joshua Puckett’s blunt line that design has historically had “the least responsibility” in the EPD triangle, and AI is forcing more designers to pick up work that used to sit with PMs and engineers.
Designers are quietly becoming builders
One of the most striking findings is that 59% of designers built their own app, tool, or utility in the last six months, often for themselves rather than their employer. Geoco makes it vivid with a story from Superpower, where a video lead with zero coding background used Claude Code over a weekend to build a Spotify-styled Premiere Pro plugin, and with Ramp’s internal numbers: 84% active on AI tools, 800 builders, and 1,500 apps shipped in six weeks.
The blockers are time, overwhelm, and trust
When designers aren’t adopting, the top reasons are tightly clustered: lack of time (57%), too many tools to evaluate (53%), and worries about output quality (52%). Geoco says many people feel “too employed” to experiment, while the wider design world is also wrestling with a deeper question: where does the tool stop and the craft begin?
The happiness gap is real, even if the causation isn’t settled
The emotional center of the video is the satisfaction data: designers doing no AI coding reported 5.93/10 satisfaction, while heavy vibe coders came in at 7.39/10. Geoco is careful not to overclaim causation, but he clearly didn’t expect the gap to show up so cleanly, especially as researchers report the highest anxiety and Anthropic’s own randomized trial found AI-assisted learners scored 17% lower on mastery.
His warning isn’t “panic,” it’s “don’t deny what’s happening”
Geoco goes out of his way not to shame the 37% of holdouts, acknowledging real constraints like enterprise lock-in, weak outputs, and teams still stuck at the lunch-and-learn stage. His concern is narrower and sharper: if you reject that this shift is happening across the industry, and reject that you may now be in the minority, you risk getting shut out of a field you still love.