I am interested in how people learn, participate, and exercise judgment in data-rich and AI-mediated environments. My research focuses four areas: Augmented Expertise (How can we cultivate specialized expertise among distributed, non-expert participants through adaptive scaffolding and hybrid human–AI systems?), Hybrid Intelligence Systems (How can humans and AI systems learn together in ways that improve both human capability and machine performance?), Algorithmic and Data Justice (How can data-driven systems in the public sector operate equitably and reflect community values?), and Civic Data and Participation (How can communities meaningfully participate in data-driven governance and decision-making?)

This research explores how people develop domain-specific expertise in complex environments without formal training. Drawing on work in citizen science systems such as Gravity Spy, I examine how interfaces, feedback, and structured participation enable non-experts to engage in sophisticated analytical work and contribute to scientific discovery.

The diagram for Gravity Spy data analysis
Topics: crowdsourced learning environments; language socialization in online communities; motivation and participation in citizen science
Contributions: this work has contributed to understanding how expertise emerges through participation rather than formal training, introduces language adoption as a measurable mechanism of socialization and learning, and shows how structured task design and feedback loops support skill development in distributed systems
Representative Work

This work moves beyond traditional pipelines where humans simply label data for AI systems. Instead, I study co-learning environments where humans and algorithms iteratively improve one another, forming adaptive systems that evolve over time.

Gravity Spy system interface
Topics: citizen science and machine learning integration; machine-guided classification systems
Contributions: this stream of search introduces co-learning as a model for human–AI interactions; shows how humans can act as more than data labelers—serving as analytical collaborators for machines; and demonstrates how hybrid systems improve both model performance and human expertise.
Representative Work

This research examines how AI systems embed assumptions about fairness, legitimacy, and decision-making. It develops participatory approaches in which communities act as evaluators of algorithmic systems, bringing contextual knowledge and normative reasoning into oversight processes.

The interface for AI learning
Topics: participatory AI auditing; public perceptions of data use
Contributions: this work reframes AI auditing as a participatory and socially grounded process; demonstrates the importance of normative reasoning in evaluating algorithmic systems; and bridges technical AI evaluation with community-centered perspectives on responsible AI practices
Representative Work

This work focuses on inequities in access to data and participation in civic processes. I study how environmental and public-sector data systems can better represent community knowledge, particularly in contexts of environmental justice and local decision-making.

The interface for Knowledge Map
Topics: environmental justice mapping; civic data representativeness; Knowledge Map (community data portal)
Contributions: this work identifies structural barriers to participation in civic data systems; develops models for understanding representativeness in community-generated data; designs systems that integrate community narratives with institutional data (See: Knowledge Map); designs systems that support third-sector knowledge of legislation
Representative Work