Federal Project Grant Award Summary

Vanderbilt University received a $397,084 Project Grant from the National Science Foundation's Division of Information and Intelligent Systems under the Computer and Information Science and Engineering (CISE) program (CFDA 47.070) on June 1, 2026, for a five-year research initiative concluding May 31, 2031. The award funds a comprehensive research project titled "CAREER: Enabling Human-Guided AI for Code" that develops advanced artificial intelligence tools for software engineering by integrating human cognitive patterns into AI model design. The research deliverables include empirical studies measuring programmer attention during software engineering tasks using eye-tracking technology, a cognitive simulation framework based on the Adaptive Control of Thought-Rational (ACT-R) architecture to generate large-scale human attention datasets, and incorporation of human attention signals into graph neural networks and transformer-based AI models to improve code generation, summarization, and debugging capabilities.

This CAREER award produces both foundational research contributions and practical AI tools designed to enhance software reliability, security, and developer productivity. The project's significance lies in advancing explainability and trustworthiness of AI systems by grounding their development in human expertise, ultimately strengthening the nation's capacity for technology innovation. The research spans multiple technical domains, combining human factors research with cognitive modeling and state-of-the-art machine learning methodologies to create AI-assisted programming tools that more closely align with how experienced software engineers analyze and reason about code.