Deployed AI agents — built for real classrooms
Each agent is practitioner-built using the PCTII cycle, classroom-tested, and grounded in published frameworks. These are not prototypes — they are working tools used by real students and educators, governed by the Educator-in-the-Loop (EITL) architecture.
Simulated practice in fictional management and leadership case studies. Deployed in EM6000 at Western Michigan University, this agent guides students through complex team dynamics, conflict resolution scenarios, and management decision-making in a simulated environment — building skills that transfer directly to real workplace contexts.
Management and leadership cannot be learned through passive instruction alone — students need to practice decision-making under realistic conditions. This agent creates a low-stakes simulation environment where students can apply, test, and reflect on management concepts before encountering them in the workplace. It supports experiential learning while keeping the educator in control of the pedagogical boundaries through the EITL architecture.
Guiding students in reflecting and brainstorming for projects. Trained on the Team Dynamics and Conflict Resolution (TDCR) module, Project Pal assists students with project planning, task assignment, conflict resolution, and team dynamics — while scaffolding their thinking rather than replacing it. Supports the LTTR framework by guiding transparent, traceable AI use.
Team conflict is one of the most common barriers to student success in project-based learning — and one of the least addressed. Project Pal gives students access to structured guidance exactly when they need it, without waiting for instructor availability. Critically, it is designed to guide reflection rather than provide answers, preserving the productive struggle that builds genuine competency in team dynamics and conflict resolution.
Creates personalized, scaffolded story problems for undergraduate engineering curriculum. Uses a structured A–L intake checklist to constrain AI output within pedagogical boundaries — ensuring generated problems are educationally sound, contextually rich, and aligned with learning objectives. Embeds the EITL two-layer governance architecture and PCTII cycle. Underpins the accepted ASEE 2026 paper on story-driven AI-augmented instruction.
Abstract engineering problems lose students who cannot connect them to real contexts. Story-driven problems increase engagement, support contextual learning, and make technical content accessible across different student backgrounds. The A–L intake checklist ensures every generated problem is pedagogically sound before a student ever sees it — AI serves the learning objective, not the other way around.
Supports faculty in integrating trauma-informed pedagogy and pedagogy of care into their teaching practice. Developed to accompany the presentation Pedagogy of Care: Teaching Through Grief, Trauma, and Vulnerability at the Magna Faculty Development Professionals Conference (August 2025). Built using the PCTII cycle, this agent reflects a commitment to designing AI tools that serve the full humanity of educators — not just their instructional efficiency.
The agent is grounded in peer-reviewed scholarship: Teaching Through Grief, Designing for Care: Toward Trauma-Informed and Student-Centered Pedagogy (Aref & Paul, 2026, Higher Education Research & Development, 45(2), 366–373). The paper argues that grief can catalyze pedagogical transformation and that emotional vulnerability fosters resilience, empathy, and deeper engagement — proposing a student-centered, trauma-informed model of teaching that positions classrooms as spaces of restoration.
Read the paper → DOI: 10.1080/07294360.2026.2617298Faculty developers often understand trauma-informed pedagogy conceptually but struggle to translate principles into concrete classroom decisions. This agent bridges that gap — providing structured, on-demand support grounded in care ethics and grief pedagogy. It models the same values it teaches, demonstrating that AI tools can serve human flourishing rather than replace human judgment.
A specialized instructional design collaborator built to bridge the gap between STEM expertise and advanced pedagogical theories including Universal Design for Learning (UDL), Backward Design, and the Ethic of Care. Functions as a pedagogical assistant, tracking the evolution of course ideas while providing evidence-based suggestions grounded in the works of Dewey, Noddings, and Gagné. Used to evaluate, develop, and refine every aspect of the course lifecycle to ensure all content meets Quality Matters standards while remaining ethically sound and educationally transformative.
Course design is one of the most complex intellectual tasks faculty undertake — and one of the least supported. This agent provides structured, theory-grounded guidance through every stage of the course development process, ensuring that pedagogical decisions are intentional, evidence-based, and aligned with both learning outcomes and ethical principles. It extends the educator's capacity without replacing their judgment.
A structured, educator-designed prompt library developed for an online people-management course. Offers students a principled middle ground between unrestricted AI use and complete restriction — six curated prompts that scaffold ethical thinking rather than replacing it, and support learning rather than undermining it. Grounded in a proposed typology for generative AI in teaching and learning.
