2Nong – Digital Farming Assistant

My Role

As the Product Designer, I led the end-to-end design process, collaborating closely with AI engineers, product managers, and developers to deliver a scalable and user-centered solution.

• End-to-end Product Design (0 → 1)
• UX Research & Usability Testing
• Interaction & Visual Design
• Design System & Documentation

Opportunity

Translate AI intelligence into a guided, trustable, and low-friction experience.

• Reducing cognitive load
• Guiding users step-by-step
• Building trust through clarity and feedback

Research & Insights

We conducted user interviews and field observations with farmers.

User personas

Mr.Tu

Age 60

Rural • Low digital literacy

Relies on experience and visual observation to identify crop issues. Struggles with digital tools and prefers simple, guided interactions.

• Visual-first understanding
• Needs quick answers
• Low familiarity with apps

“I just want to know what’s wrong and how to fix it.”

Ms.Ba

Age 32

Active user • Social learner

Actively engages in farming communities to ask questions and share knowledge. Values trust and peer validation.

• Learns through community
• Trusts peer recommendations
• Wants better decisions

“I trust what other farmers say more than generic advice.”

Mr.Bay

Age 48

Tech-savvy • Data-driven

Uses data and tools to optimize farming outcomes. Interested in market trends, pricing, and scalable insights.

• Needs market insights
• Comfortable with technology
• Focus on efficiency

“I need reliable data to make the right decision.”

These personas informed key design decisions, leading to a visual-first, guided experience that reduces complexity and builds user trust.

• Farmers rely heavily on visual recognition, not text
• Trust increases with clear actions
• Users prefer guided flows
• Too many options → abandonment

Solution

Based on user behavior and needs, we designed a guided, low-friction experience that enables farmers to quickly identify crop issues and take immediate action:

• Capture crop image
• AI processes
• Instant diagnosis
• Treatment recommendation

Design Approach

We translated user insights into a design approach that simplifies complexity, reduces cognitive load, and enables scalable product growth:

• Simplified AI outputs into human-readable insights
• Reduced cognitive load
• Built scalable system

Design System

We developed a scalable design system that aligns design and engineering ensuring consistency, reducing implementation gaps, and enabling faster product iteration across App and Web.

Transformation-driven

The final UI transforms complex agricultural data and AI outputs into clear, human-centered experiences bridging the gap between technology and real-world usability.

Onboarding & Login

The onboarding experience uses familiar farming visuals to build immediate understanding and trust, reducing the barrier for first-time users.

The login flow prioritizes speed and accessibility by using phone number-based authentication a familiar and convenient method for users reducing input effort and enabling quick access to the product.

AI-Powered Crop Disease Diagnosis

The AI-Powered Crop Disease Diagnosis experience enables farmers to quickly identify crop diseases through a guided image capture process, ensuring accurate and reliable input.

The system translates complex AI outputs into clear, actionable results including disease identification, visual highlights of affected areas, and step-by-step treatment recommendations helping users take immediate and confident action.

Market Insights

The market insights experience transforms complex pricing data into clear, easy-to-understand information allowing farmers to quickly grasp market trends.

With localized pricing and visual indicators, users can make timely and informed decisions on selling, purchasing, and planning their agricultural activities.

Impact

The final product delivered measurable improvements in speed, accuracy, and user confidence helping farmers make faster and more informed decisions.

40%

Faster diagnosis time

60%

Improvement in early detection

80%

Completed flow <30 seconds

Reflection

This project highlighted a critical shift: success is not determined by how intelligent the system is, but by how clearly users can understand and act on it.

For farmers, effective design means:

• Clarity over complexity
• Guidance over information overload

Next Steps

Looking ahead, we focused on scaling the ecosystem and improving accessibility by addressing key user needs and product gaps:

• Offline mode for rural areas
• Voice interaction for elderly users
• Community knowledge sharing
• AI confidence visualization (highlight affected areas)