Skip to content
Keythe Rueckert

30-second summary

Keythe Rueckert, Product Designer

Product Designer who builds in code. I design workflows for complex B2B software, then prototype them in React and TypeScript.

Experience

  • Design Engineer, Cantu Inc. (Aug 2026 to present). Building interfaces between product design, design systems and frontend engineering, working directly in code.
  • Product Designer, Cantu Inc. (Nov 2024 to Aug 2026). Product designer across B2B CRM, supply-chain and e-commerce products: research, prototyping in Figma and front-end validation.
  • Earlier: UX/UI Designer at Focus Monk (Sep 2023 to Nov 2024) and Lize (Apr to Aug 2023).
  • Postgraduate, UX Engineer, PUC Minas (in progress, Aug 2026 to Dec 2027).

Working with me

Availability
Open to full-time and contract roles as a Product Designer who builds in code, or as a Design Engineer.
Location
Brazil (UTC−3), working remotely.
Languages
English (advanced, professional working proficiency). Portuguese (fluent).

Best work

  1. Fewer how-to tickets by putting help where the question starts

    About 20% fewer tickets overall. I did not build the assistant; I owned the knowledge it answers from. No user test yet.

    Open the case: Fewer how-to tickets by putting help where the question starts

  2. Prototypes that obey the design system, and say what they could not check

    Illustrative figures: 61% fewer review rounds with engineering. First version working, pilot not run yet, no adoption data.

    Open the case: Prototypes that obey the design system, and say what they could not check

  3. A design system you can read at a glance

    I built a health dashboard and two pipelines so analysts and developers work from the same real catalog. It is tested and the team is starting to adopt it; I have no adoption numbers yet.

    Open the case: A design system you can read at a glance

How the work is checked

The help center is real work with the prototype rebuilt in code and sample data; the verified prototypes are a recreation with fictional data; the third is an excerpt of an internal tool. Each case says what was measured and what was not, including where there is no user research or adoption data.

Stack

  • React
  • TypeScript
  • Next.js
  • Tailwind CSS
  • shadcn/ui
  • TanStack Table
  • Storybook
  • Vitest
  • axe-core
  • Figma

Contact