PROJECT:

Bachelor's Thesis (B. Sc. User Experience Design)

TOPICS:

UX Research, UI Design & Branding, Frontend Developoment

DURATION:

June – November 2024

TOOLS:

Figma, React, Tailwind CSS, Miro, Canva, Vite

TESTING

Cognitive Walkthrough + Remote Usability Testing (n=7)

Cover
Cover

Feldschau

A User-Friendly Platform for AI-Based Plant Stock Evaluation

With rising food insecurity and a growing world population, plant breeders need efficient tools to assess crop performance.

Feldschau.AI is a web platform designed to support AI-powered evaluation of wheat spike data, helping breeders make faster, data-driven decisions — without replacing hands-on fieldwork.

The challenge was to make advanced AI technology accessible to a non-tech audience in agriculture.

solution.

Feldschau.AI provides:

  • A dashboard-based platform for breeders to manage trials and view results

  • AI-generated visual feedback on images of crop trials

  • Support for trial data management, image uploads, and score filtering

  • A responsive design tailored for tablet use in the field

Showcase image
human-centered design.

Through a user-centered process, I interviewed experts, developed detailed personas (e.g., "Markus – the senior breeder"), and derived critical needs: simplicity, transparency, and efficiency.
The platform helps breeders evaluate, compare, and export plant trial data - while keeping them in control.

Showcase image
process.
  1. Research: Competitor analysis & expert interviews

  2. Personas & User Requirements: Needs translated into concrete features

  3. Concept Development: Scribbles → Wireframes → Lo-Fi prototype

  4. UI & Branding: Moodboard, design system, and Hi-Fi prototype in Figma

  5. Usability Testing: Remote testing with real plant breeders (Reddit community)

  6. Implementation: Developed frontend in React + TailwindCSS, using localStorage for demo functionality

result.

While breeders are open to AI, they demand transparent results and prefer tools that support — not replace — their expertise. Human-AI collaboration is key.

Outcome:

  • Fully implemented interactive prototype (React frontend with simulated data)

  • Demo video showcasing user flow, core features, and interface design

  • SUS Score: 84.64 (Excellent usability)

  • Design praised for intuitive flow, responsive layout, and clear data visualization

  • Usability insights led to improvements in filtering, feedback clarity, and accessibility

more to explore.