Stephen Haney has seen this movie before—designer builds tool, developers ignore it, tool languishes. Which is why his second act, a San Francisco startup called Paper, is making a different wager entirely.
The company announced a $34 million Series A round on July 23, co-led by Accel and ICONIQ, with backing from Designer Fund and a collection of builders whose day jobs involve wrangling AI: Michael Grinich of WorkOS, Anton Osika of Lovable, and engineers from Anthropic and OpenAI. It's the kind of cap table that suggests Paper isn't just building software—it's building for a workflow that doesn't quite exist yet.
Founded in 2024, Paper positions itself as a "design platform for the agentic era." Strip away the jargon and what's left is this: a web-based canvas that outputs HTML and CSS directly, letting designers work side-by-side with AI coding agents like Claude Code or GitHub Copilot without the usual translation headaches. No hand-off, no export, no friction. At least, that's the pitch.
When Accel doubles down
Accel already led Paper's $4.2 million seed round back in February 25, 2025. That the firm is writing another check fifteen months later says something. In its investment memo, Accel framed the bet around a shift it sees coming—away from traditional design-to-development handoffs and toward real-time collaboration between humans and agents. ICONIQ, for its part, calls Paper "infrastructure for the agentic era," a phrase that manages to be both sweeping and vague.
Paper's desktop product went live in late February 2026. The early customer roster reads like a who's-who of the developer-first startup world: Ramp, Vercel, PostHog, Quartr. Y Combinator itself is reportedly using the platform, though whether that constitutes an endorsement or simply proximity is hard to say.
A second swing at a familiar problem
Haney's credibility here isn't incidental. He co-founded Modulz, another stab at web-native design tooling, and built Radix UI—an open-source component library now embedded in frameworks like shadcn. WorkOS acquired Modulz and the Radix team in June 2022, which means Haney has already navigated the treacherous waters of building tools developers might actually use. That experience matters in a space littered with beautiful design apps that engineers quietly route around.
His track record suggests he understands the fundamental tension: visual tools that produce code developers won't touch are just expensive wireframing apps.
What the money buys
The $34 million will flow toward product development, hiring (LinkedIn currently lists Paper's headcount somewhere between 11 and 50 employees), and refining those AI-native workflows. Paper's positioning explicitly contrasts its HTML/CSS rendering with Figma's proprietary WebGL approach—a technical choice that doubles as a philosophical argument about where design tools are headed once agents can actually read and modify the canvas.
Valuation remains undisclosed. In a June 2026 LinkedIn post, Haney claimed the company was already profitable, though that remains unverified and, for an early-stage startup pulling in venture capital, perhaps beside the point. Profitability is nice; market positioning is everything.
The real question, of course, is whether the "agentic era" Haney and his investors are betting on arrives on schedule—and whether designers will trust AI agents enough to let them rewrite their work in real time. That's a cultural shift as much as a technical one. But then, someone has to build the tooling before the culture can catch up.
Or maybe Paper is early. The optimistic read is that Haney's been here before and knows the timing better this time around. The skeptical read is that being right too soon looks a lot like being wrong.
