Using GenAI to deliver on climate goals
Why this mattered
International buyers face increasing pressure to achieve ambitious climate goals, but their supply chains remain fragmented and difficult to transform at scale.
In Vietnam’s coffee supply chain, large multinational corporations struggled to transition thousands of smallholder farmers to low-carbon, regenerative practices. Each farm had unique conditions, farmers lacked technical expertise, and traditional extension services couldn’t scale.
Without direct, tech-enabled support reaching every farmer, climate commitments remained aspirational rather than actionable.
We were brought in to design and deploy an AI-first solution that could deliver personalized agronomic guidance to remote farmers at scale.
Approach
We embedded ourselves in the buyers’ existing training programs to understand farmer workflows, constraints, and digital literacy levels. Rather than building a complex system, we designed a mobile-first experience that abstracted agronomic complexity into something any farmer could use.
We delivered a production-grade AI platform in partnership with agronomists and sustainability teams.
- We conducted field research with farmers and agronomists to understand real-world constraints and adoption barriers
- We designed a mobile-first interface requiring minimal digital literacy. Farmers simply take photos of their farms, crops, and fertilizers
- We built a multi-agent AI orchestration system that analyzed farm conditions, synthesized agronomic best practices, and generated personalized transition plans
- We abstracted the complexity of regenerative farming practices, making expert guidance accessible to every farmer regardless of education level
- We integrated with buyers’ existing training programs and field extension services for seamless adoption
- We provided agronomists with a dashboard to monitor adoption, validate AI recommendations, and intervene when needed
What changed
The pilot exceeded expectations, with 70% adoption rate among participating farmers thanks to the platform’s ease of use and relevance.
Farmers received actionable, personalized advice tailored to their specific farm conditions, something impossible to deliver at scale through traditional extension services.
The buyer gained visibility into farm-level transitions and could credibly track progress toward climate commitments.
- 70% adoption rate among farmers in the pilot program, far exceeding typical digital agriculture uptake
- Farmers reduced input costs through optimized fertilizer usage while improving soil health
- The buyer could track and verify regenerative practice adoption across their supply chain
- Agronomists scaled their impact from dozens to hundreds of farms without proportional headcount growth
- The platform created a foundation for carbon credit verification and climate reporting
The platform abstracted complexity through a simple user flow:
- Farm assessment: Farmers photograph their crops, farm, and current fertilizers
- AI analysis: Multi-agent system analyzes images, fetch farmers data from buyers’ databases, finds the relevant agronomic advice bsed on the crop development stage, and identifies the next potential nudge to move farmers towards regenerative farming
- Personalized guidance: Farmers receive step-by-step recommendations for transitioning to regenerative practices, based on their own farms and crop development stage
- Progress tracking: Both farmers and agronomists monitor adoption and measure impact over time
- Monitoring: The system generates reports that can be leveraged later for carbon verification
The AI orchestration layer handled complex agronomic reasoning:
- Crop stress and disease detection
- Derived crop development stage from a combination of data and photos
- Fertilizer optimization based on current usage and farm conditions
- Regenerative practice sequencing tailored to farm readiness
- Cost-benefit analysis for farmer decision-making
Could this apply to your organization?
If you make mission-critical decisions based on prior research & analysis, you need a proper AI intelligence infrastructure
Typical examples include:
- Professional services firms: Scaling advisory capabilities through AI-assisted diagnostics, generating tailored recommendations while maintaining institutional-grade oversight.
- Financial services & portfolio management: Providing structured, AI-assisted due diligence guidance across portfolio companies, standardizing reporting, or embedding expert judgment into scalable review workflows.
- L&D and training programs: Scaling expert support and training across fragmented suppliers, enforcing central SOP while maintaining a high degree of oversight on accuracy and uptake.
How you can get started
We typically help organizations:
- Assess readiness for AI-enabled supply chain transformation
- Identify high-impact intervention points where personalized guidance drives behavior change
- Design AI-first experiences appropriate for target user digital literacy and connectivity
- Build AI orchestration systems that embed expert judgment and adapt to local conditions
- Integrate with existing programs (training, certification, incentives) for seamless adoption
- Establish monitoring and validation frameworks to build trust and ensure quality
- Move from pilot to scaled deployment across supply chain networks
