Optimizing Nonprofit Grants with Generative AI

Optimizing Nonprofit Grants with Generative AI

Case Study

 

Optimizing Grants with Generative AI

Customer confidential

Harnessing the power of generative AI, this non-profit customer teamed up with SDG to enhance its grant analysis process.

Challenges

Latency in data retrieval and analysis
Understanding and implementing the generative AI Stack
Disparate data formats and sources

SDG Solutions

Generative AI stack with Large Language Models (LLM), LangChain, and LangGraph
Centralized vector store & data lake in AWS (Amazon Web Services)
Multi-agent orchestration for enhanced grant analysis

Improving grant evaluation

Our customer is a national nonprofit dedicated to improving access to higher education and career success among underserved populations. Through strategic grants and program-related investments, their foundation supports innovative solutions that increase college success and career readiness.

Since May of 2015, this foundation has invested over $300 million in grants to nonprofit organizations that expand educational opportunities and foster upward mobility for underserved communities. To enhance the speed and accuracy of their grant evaluation process, they partnered with Solution Design Group (SDG) to explore using generative AI to identify patterns and provide actionable insights in real-time. By addressing challenges like latency and data relevancy, SDG and the customer laid the groundwork for a responsive, AI-enhanced grant analysis workflow.

Streamlining grant analysis

This foundation’s commitment to expanding educational access is supported by its rigorous grant evaluation process, which relies on deep analysis to make high-impact funding decisions. With over $300 million in grants awarded, the foundation faced increasing demands for accurate, real-time insights to guide its investments. Recognizing the potential of generative AI and Retrieval Augmented Generation (RAG) to meet these needs, they collaborated with SDG to prototype a comprehensive generative AI technology stack that would accelerate the grant analysis process while ensuring that recommendations were data-driven and contextually relevant.

With a wealth of historical grant data and real-time application information, the foundation was well-positioned to leverage RAG-powered LangChain agents to process and analyze data efficiently. SDG’s implementation allowed them to go beyond conventional analysis, gaining real-time insights and retrieving information directly related to each grant’s objectives, applicant history, and outcomes.

ECMC case study inline image 2

The right approach

Leveraging the full generative AI stack

SDG’s consultants recommended that the customer use a generative AI technology stack, comprising AWS, LLMs, and LangChain agents. This approach included AI models that retrieved relevant information, generated summaries, and offered actionable insights tailored to each grant’s criteria. SDG worked with them to orchestrate a collection of LangChain and LangGraph agents to rapidly analyze incoming grant and demographics data, reducing the need for manual data processing and allowing analysts to focus on high-impact evaluation and decision-making.

  • Real-Time Data Retrieval and Adaptive Analysis
    LangChain and LangGraph automated agents could quickly retrieve and analyze information from AWS data sources using Retrieval Augmented Generation (RAG), providing the foundation’s grant analysts with immediate access to applicant insights and grant impact metrics.
  • Dynamic Recommendation Engine
    LangChain Agents used generative AI to identify potential risks and assess a program’s alignment with the foundation’s mission, ensuring each decision was backed by robust, real-time data.

A new standard in grant analysis

A significant challenge the team faced was orchestrating the integration between different tools to create an efficient pipeline while minimizing latency and hallucinations. Additionally, the complexity of retrieving real-time data through RAG while maintaining conversational coherence presented unique challenges for engineering prompts while optimizing response times.

Using LangChain and LangGraph autonomous agents helped the conversation flow more coherently, allowing for complex decision-making paths within interactions. Modeled after a multidisciplinary team of subject matter experts, this multi-agent architecture was structured as a graph of agents, each specializing in a different aspect of the workflow. LangGraph’s directed graphs enabled flexible routing of tasks between agents based on the user’s input and the context of the conversation.

Agents were also given power to generate SQL and retrieve relevant demographics and grant data. By augmenting the LLM’s inherent knowledge with external, up-to-date data sources, the agentic bots were able to deliver precise and contextually relevant results.

By embracing the full generative AI stack, the foundation has set a new standard in grant analysis and reporting, allowing for real-time, data-rich insights and adaptive decision-making. This transformative approach to grant analysis puts this foundation in a position to expand its impact supporting nonprofits that deliver critical educational opportunities and enabling faster, more effective funding decisions.

Reducing global methane emissions (company confidential)

Reducing global methane emissions (company confidential)

Case Study

 

Reducing global methane emissions

Global climate advocacy organization (company confidential)

A global non-profit climate advocacy organization partnered with a marketing and strategy firm and SDG to transform their complex methane emissions calculation tool into a sophisticated, easy to use, and scalable cloud-based web app.

Challenges

Complexity of the solution
Geopolitical nuances
Deep industry data
Diverse policy content sources
Globally distributed stakeholders

SDG Solutions

Product mindset
Product leadership
AWS Cloud Architecture
Software engineering

Using data to develop effective global climate policy

Our customer is a world leading climate advocacy organization working to advance technology and policy changes needed to achieve a zero-emissions, high-energy planet at an affordable cost. Methane emissions are 80x more damaging than carbon dioxide and reducing them is critical for our planet to stay within the 1.5 degree warming threshold. The organization works with governments, industry leaders, and other NGO’s across the globe to affect deep, global reductions in methane emissions to protect against the risk of irreversible changes to our climate.

To that end, they developed a Methane abatement tool, which allows countries to record and continually refine their emissions estimations based on the best available information about a country’s oil and gas industry. By using the emissions and inventory calculations in the tool, countries can explore variables and specific policy options that can result in pollution reduction. While the original spreadsheet-based solution was impressive, it was inherently complex, and not easy to use, maintain or scale simultaneously across multiple country engagements. A longer-term, more user-friendly solution was necessary and the customer engaged with their marketing partner and SDG to build it.

An ambitious vision

As a strategic partner to the organization, the marketing firm conceived a winning product strategy to transform the existing tool from a heavily manual process into a powerful web- and cloud-based software application. Turning this vision into reality required deep technical knowledge, so they engaged SDG to form a strong product development team. Together with the customer’s policy and industry experts, the team developed a state-of-the-art user experience, and an innovative, data-driven approach for the new solution, transforming it into a dynamic, adaptable and accessible platform.

SDG and the marketing and strategy firm worked to deliver a scalable product strategy and design a technical architecture using a combination of third-party tools and custom software. As part of the design and system development work, and after comprehensive review of various platforms, the team’s architects and designers agreed that an AWS-based solution would provide the combination of scalability, user experience and performance required for the new application and its mission-critical purpose.

A scalable solution

The new application runs on an AWS cloud-based, 3-tier, service-based architecture that was tailored to the needs of the organization’s globally distributed staff and user base. The user experience layer comprises web pages, forms, data visualizations, and content repositories custom-designed by the product team and delivered through a CMS and other integrated front-end tools running on Amazon Web Service (AWS).

While this organization has a global impact, they have a limited IT infrastructure capability and staff, and the cost of building and managing an on-prem solution and data center would prove untenable. With the AWS technology that the team and SDG implemented, the application can run on a serverless architecture that scales automatically to handle load from users across the world. Because of this architecture, the customer’s limited IT staff can offload management of the system to the cloud, while their analysts, researchers, and policy advisors can confidently collaborate with various country stakeholders using a powerful, modern web application.

CATF case study inline image

Tackling climate change with AWS

In 2023 the pilot of the new application was launched in their AWS environment, and team members have begun to import climate and policy data into the new system. The new application will accelerate the ability to forecast and establish abatement targets and recommend mitigation plans and policy changes for countries using the platform – which in turn will result in targeted impact and reduced methane emissions from the oil and gas industry across the globe.

The modernized version of the tool has allowed them to expand their engagement and influence. With their modernized application up and running, this customer is now poised to work with more countries and institutions, develop more comprehensive policy solutions, and accelerate their efforts to decarbonize the global energy system.

Second Harvest Heartland: Mapping Food Insecurity with Big Data

Second Harvest Heartland: Mapping Food Insecurity with Big Data

Case Study

 

Mapping Food Insecurity with Big Data

Second Harvest Heartland

A leading food bank worked with SDG to use data on hunger to build interactive maps to track food insecurity.

Challenges

Non-profit dynamics
Disparate data sources
Challenging visualizations

SDG Solutions

Technical design
Data visualization
User experience design

Attacking hunger through sophisticated data visualization

Second Harvest Heartland is one of the nation’s largest, most efficient, and most innovative food banks. In a typical year, the organization collects, stores, and distributes over 80 million pounds of food to an estimated 530,000 people in 41 counties in Minnesota and 18 counties in western Wisconsin.

Second Harvest’s mission is to end hunger through community partnerships. They achieve this by focusing on results, innovation, and thought leadership. They find new sources of food and deliver it to over 1,000 food shelves, pantries, and other agency partner programs that in turn distribute this food to hundreds of thousands of families.

One way Second Harvest achieves this mission is through educating community leaders and legislators about food insecurity levels and associated demographics in their areas. They use complex demographic and geographic data to tell this story, but this raw data only paints a partial picture and it’s often difficult to visualize. They needed a more engaging and effective experience to communicate this information.

An interactive solution to tracking hunger

Second Harvest partnered with SDG to conceive, design, and implement an interactive map that allows users to navigate their service area and visually highlight hunger need areas. The map helps Second Harvest represent and understand hunger needs more accurately, and communicate with stakeholders to enact change and generate funding for more outreach.

An interactive experience compatible with any device

After collaborating with Second Harvest to envision and research the solution, SDG developed an interactive map that both looks and functions well across any device. This responsive web application is able to recognize, organize, and scale its display to the device of the user.

The interactive map is a single-page application, where all of the code needed for the page to function is loaded once. As the zoom level, demographic filters, and other data are adjusted, the interactive map dynamically communicates with the backend data and automatically redraws itself. This user interaction is seamless and fluid providing a fast and delightful user experience.

User controlled, powerful, and information-rich

The data in the application is detailed enough to display hunger needs by census tract, or clusters of about 8,000 people, which allows the user to efficiently view hunger need hot-spots within a community.

Additional user-controlled data layers provide useful information on items such as food shelf locations, the amount of food distributed at each location, Commodity Supplemental Food Program (CSFP) locations, school locations, students receiving free or reduced meals, and poverty rates. The application is a major step forward in communicating targeted food insecurity data across a population.

Partnership scorecard

The interactive hunger map helps Second Harvest execute its mission of attacking hunger in Minnesota and Western Wisconsin.

Pounds of food annually

Counties mapped

People served

To read more success stories, visit the our work page.

Photo by Joel Muniz on Unsplash