.png)
Transforming Food Waste Recovery with Generative AI
Protagona partnered with a leading food recovery platform to integrate generative AI into their volunteer coordination system, automating updates across thousands of food banks and helping fresh produce reach communities faster and at greater scale.
Industry
Nonprofit
Teams & Services
AI/ML Engineering, Back-End Development, Data Engineering
Tech & Tools
Generative AI, AWS (Lambda, Bedrock, DynamoDB, EventBridge, SQS), Google Vertex AI, Contact Management Systems
Key Data Points
The Vision
In a world where millions face food insecurity while tons of fresh produce go to waste, a pioneering food recovery organization envisioned a future where technology could bridge this gap more effectively. The organization recognized that their existing volunteer coordination system, while functional, could be dramatically enhanced through the power of generative AI to create more meaningful connections between food donors and recovery volunteers.
This vision centered on automating the process of maintaining database information, proactively finding updated contact information, web addresses, locations, and hours of operation of over 8,000 food banks, enabling more efficient food recovery operations across communities nationwide.
The Goal
The primary objective was to revolutionize the organization's contact management system by integrating generative AI capabilities that could intelligently process, organize, and utilize volunteer and donor contact information. The organization sought to leverage AI to reduce the manual overhead of coordinating between food donors, volunteers, and recipient organizations, while ensuring that fresh produce reaches those in need more efficiently and with greater scale. It also lightened the load on donors and food banks by autonomously updating food bank information, eliminating the need for manual information gathering or tedious, repetitious contact between representatives.
The Challenge
The existing contact management system presented several significant challenges that hindered optimal food recovery operations. Manual processing of volunteer information and donor details created bottlenecks that slowed response times when fresh produce became available for recovery. Changes to food bank information could go unnoticed, preventing those in need from finding the resources readily available to them.
The organization struggled with efficiently matching volunteers to opportunities based on location, availability, and capacity. Additionally, the complexity of managing diverse contact information formats, communication preferences, and volunteer skill sets made it difficult to maintain accurate, up-to-date records. The system needed to handle the dynamic nature of food recovery operations, where timing is critical and opportunities can emerge with little notice.
The challenge was compounded by the need to maintain data privacy and security standards while creating a more intelligent, responsive system that could scale with the organization's growing impact across communities.
The Solution
The team designed and implemented a comprehensive generative AI solution that transformed how contact details and volunteer information are discovered, processed and utilized. The solution centered on developing algorithms that could intelligently collect, parse, categorize, and enhance contact information while providing contextual insights for better volunteer coordination.
The technical approach involved integrating Generative AI to search the internet, leveraging both existing information and Vertex AI, to collect and process the most up-to-date information for all of the organizations across the country that exist in the database. These results are then compared to existing information, and any discrepancies or updates are formatted into an easy to use form containing suggestions and sources of truth.
A key component of the solution involved creating intelligent automation workflows that could process incoming volunteer applications, donor registrations, and coordination requests with minimal manual intervention. The team implemented robust data validation and enrichment processes that ensure contact information remains accurate and actionable.
The partnership approach emphasized close collaboration with the organization's operational team to ensure the AI enhancements aligned with real-world food recovery workflows and volunteer coordination needs, creating a system that amplifies human impact rather than replacing human judgment.
.png)