The internal AI tool that's transforming how Stripe designs products | Owen Williams

2026-05-04 · Source: How I AI · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Emerging Technologies & Innovation · Depth: Intermediate, extended

Summary

Owen Williams, a design manager at Stripe, developed an internal prototyping tool called Protodash to address the "uncanny valley" effect of generic AI-generated designs that failed to adhere to Stripe's specific design system, Sale. Protodash, initially a React app running locally or in dev boxes, bundles Cursor rules and Sale components to generate high-fidelity, interactive dashboard prototypes. This tool enables designers and, increasingly, Product Managers, to quickly create realistic prototypes with various data states, internationalization, and business models. Williams further evolved Protodash into Protodash Studio, a browser-based platform that allows users to build and remix prototypes without local setup, leveraging dev box infrastructure and an embedded LLM for in-browser code generation and iteration. The tool also incorporates features for collaborative design reviews, direct canvas feedback for AI, and fidelity modes to indicate work in progress.

Key takeaway

For design and product leaders aiming to accelerate product development and improve design quality, consider investing in highly customized internal AI-powered prototyping tools. Your team can significantly reduce the "uncanny valley" effect of generic designs and foster a culture of rapid, high-fidelity iteration by integrating your specific design system and workflow into these tools. This approach not only empowers designers and PMs to unblock themselves but also transforms design reviews into interactive, data-rich discussions, ultimately leading to more precise and effective product outcomes.

Key insights

Internal tools precisely tailored to company culture and design systems significantly enhance prototyping efficiency and quality.

Principles

• Design systems are critical for consistent AI-generated prototypes.
• Lowering technical barriers empowers non-engineers to prototype.
• High-fidelity, interactive prototypes improve design reviews.

Method

Develop an internal, AI-powered prototyping tool by bundling design system components and LLM rules. Integrate with dev box infrastructure for browser-based access and enable direct canvas interaction for AI feedback and collaborative reviews.

In practice

• Integrate your design system into AI prototyping tools.
• Use dev boxes for accessible, pre-configured prototyping environments.
• Implement fidelity modes to manage prototype expectations.

Topics

• Protodash
• Stripe Design System
• AI Prototyping
• Internal Developer Tools
• Product Design Workflow

Best for: Product Manager, CTO, VP of Engineering/Data, Product Designer, AI Product Manager, AI Engineer

Related on AIssential

• The internal AI tool that’s transforming how Stripe designs products | Owen Williams — Internal tools precisely tailored to company culture and design systems significantly enhance prototyping efficiency and quality.
• 🎙️ This week on How I AI: The internal AI tool that’s transforming how Stripe designs products — Internal AI prototyping tools tailored to a company's design system enhance product development and collaboration.
• 🔵 Google Search now generates UI on demand — AI is rapidly transforming product development, shifting focus from coding to idea validation and enabling on-demand UI generation.
• More PMs than designers now use Stripe's internal AI prototyping tool — Empowering PMs with design tools improves early-stage prototyping and communication with design teams.
• Fragments: May 5 — AI's rapid evolution demands adaptable development practices, responsible deployment, and strategic infrastructure choices.
• How Stripe deploys 1,300 AI-written PRs per week — Stripe uses AI agents and cloud environments to automate engineering tasks and enable autonomous machine-to-machine payments.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by How I AI.