SBIR Phase II: Holistic System for Comprehensive Student Assessment

About This Grant

The broader/commercial impact of this SBIR Phase II project is to transform how educators analyze and respond to student learning by providing real-time insight into qualitative work such as essays, assignments, and lab reports at a scale that was previously impossible. The project addresses a long-standing challenge in education: the difficulty of systematically analyzing open-ended student work in a timely and efficient way. This technology applies artificial intelligence and natural language processing to help teachers and administrators access actionable insights that inform instructional decisions and improve student outcomes. This technology advances scientific and educational understanding by enabling real-time analysis of unstructured learning data, an area that has historically been difficult to scale. The initial market includes K–12 public schools, independent schools, and higher education institutions. As schools seek secure, scalable, and ethical AI tools, this technology offers a clear commercial opportunity. The platform’s ability to integrate into instructional workflows, such as Learning Management Systems, and keep data fully within institutional control offers a durable competitive advantage. By year three, the company expects to serve hundreds of schools, helping educators become more responsive in their teaching. This Small Business Innovation Research (SBIR) Phase II project implements a multi-agent artificial intelligence and natural language processing system to analyze qualitative learning data–or student work–providing educators with accurate, reliable, and actionable insights into student learning. This project builds on a system that was derisked with the support of a SBIR Phase I grant and addresses the serious problem in educational systems of assessing student learning on a large scale in a timely way. This project leverages a number of machine learning technologies including various natural language processing algorithms, transformer-based large language models, and proprietary machine learning models. These artificial intelligence and machine learning components are embedded in a full-stack web application that allows for efficient ingestion and preprocessing of student work along with a data dashboard that makes it easy for users to understand the generated reports. This Phase II project builds upon the de-risked technology from Phase I by expanding the available analysis algorithms and technologies, enhancing the full-stack web application to handle commercial workloads, and increases the integrations with existing educational technologies all to prepare LearningPulse for widescale commercial adoption in K-12 and Higher Education This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.