My research explores how AI technologies and human expertise can mutually enhance each other in organizational settings. My work can be categorized under two main themes:

Theme 1: AI Augmenting Human Decision-Making

In this theme, my research explores AI's role in enhancing learning, skill development, and productivity within organizations. Through field experiments, I examine how AI tools improve productivity and support skill acquisition for new learners across tasks of varying complexity. I also study the differential effects of AI usage patterns on experienced and inexperienced users, revealing how AI enhances decision-making while highlighting risks such as over-reliance. By employing a Partially Observable Markov Decision Process (POMDP) model, I gain deeper insights into how prolonged AI use impacts skill development and potential inertia against AI adoption, aiming to balance immediate productivity gains with sustainable skill enhancement. Additionally, to address global challenges, my work explores AI’s potential to bridge linguistic divides by proposing new approaches for non-native English speakers using AI-powered coding tools.

Theme 2: Humans Augmenting AI Decision-Making

This theme focuses on improving AI systems by embedding human insights to enhance their performance and transparency. A core aspect is integrating tacit knowledge and inferred latent themes into AI models, which enhances AI decision-making by aligning models more closely with real-world complexities. By employing Bayesian modeling and AI-enabled extraction & aggregation techniques, my research aims to make AI systems more robust and contextually aware, as validated in various field experiments.

Another key component involves creating explainable AI models, particularly for recruitment, where transparent decision-making is crucial. By utilizing a Hierarchical Attention Mechanism, I develop systems that highlight human-relevant qualifications, fostering trust in AI's role in decision processes.

Furthermore, I address the challenges of Generative AI (GenAI) by proposing hybrid frameworks that incorporate human guidance to enhance AI system planning, adaptability and effectiveness. This theme ultimately emphasizes how human expertise can shape and augment AI, making systems more reliable while preserving necessary oversight and human control.