Home | Research Projects | Publications | Join us

ExplorAI Research Group

ExplorAI is an AI engineering and systems research lab led by Dr. Vanessa Figueiredo in the Department of Computer Science at the University of Regina. Our work bridges cognitive psychology, human factors, software engineering, and human-computer interaction to design and evaluate interpretable, trustworthy intelligent systems that mirror human cognitive abilities. We develop modular software architectures that programmatically guide machine reasoning during execution, providing strict behavioral guarantees in complex, real-world deployment environments.

Interested in joining us? Visit our Join us page to learn more about current opportunities for prospective graduate students and postdoctoral fellows looking to work at the intersection of Language Engineering, Distributed Systems, and Human-AI Interaction.

Research Projects

Intelligent tutoring systems

Serving as our primary validation domain, this project addresses the safety and consistency challenges of digital education. Because learning environments feature dynamic, non-linear multi-turn interactions and demand a zero-tolerance threshold for factual inconsistencies, they require trustworthy architectural frameworks.

We translate dynamic human information-processing flow patterns and fuzzy expert decision-ladders into programmatic execution-time model constraints, building adaptive tutoring interfaces. This work bridges formal K-12 schooling, trade sectors, and higher education.

To support this line of research, we developed CogniConsole, an open-source runtime framework designed to scale inference-time control for large language model.

Research assistants:


Procedural content generation for games

This line of research explores how automated methods can create dynamic game elements—such as characters, rich lore, and dialogue trees—without sacrificing narrative coherence.

Instead of relying on unconstrained model generations, we use symbolic scaffolding and memory loops to map an NPC's background story and core personality traits directly onto execution-time state validation schemas. Additionally, we design advanced techniques for persistent conversation history, allowing agents to navigate complex state-space transitions over multi-turn gameplay without behavior degradation.

Research assistant:


Human-Centered Assistive Technologies & Vulnerable User Modeling

To ensure automated reasoning aligns safely with human cognitive load thresholds, this project models the implicit behaviors, interaction vulnerabilities, and constraints of specific user groups, including children and older adults.

Through participatory design and controlled experiments, we investigate how unconstrained data presentation disorients users during digital discovery tasks. We translate these cognitive insights into lightweight, real-time support tools and inclusive reference frameworks that protect user autonomy and reduce interactive friction.

Research assistants:


Distributed Systems & Multimedia Data Synchronization

Building on our co-patented, event-driven web-synchronization architecture (originally operationalized via Tagbly and deployed at scale at the Beaty Biodiversity Museum), this project scales standalone inference-time control layers into distributed topologies.

We are developing lightweight synchronization primitives and structural schema languages necessary to maintain behavioral invariants across decentralized, edge-native agent networks. This ensures that when unexpected user context variations occur across isolated devices, the system eliminates state fragmentation and prevents cascading execution failures.


Back to top

Publications

Access Google Scholar for a complete list of publications.


Back to top


Join us

There are no openings in our research group for the 2026 cycle.

Are you interested in developing and researching human-centred intelligent systems?

Our research group welcomes students from diverse academic backgrounds who are eager to engage in exploratory research, understand user needs, and evaluate technologies designed to solve real-world problems. Our students are curious, open to learning from users, and inspired to translate their experiences into impactful solutions.

Our work revolves around collaborating with real-life users to create intelligent systems that adapt to their needs, behaviors, and challenges. This means stepping outside your comfort zone to engage with users from various social and cultural groups, gathering insights to design interfaces and features that truly support them.

We are looking for students with a strong foundation in human-computer interaction research methodologies, human-centered design approaches, intelligent systems, or related areas.

If you're interested in joining ExplorAI, please read these guidelines.


Back to top