🏆MerryQuery won Best Technical Demonstration Award at AAAI 2025!

AI-Powered Learning for the Modern Classroom

MerryQuery is an AI-powered educational assistant that uses Retrieval-Augmented Generation (RAG) to provide course-specific, contextually relevant responses. With guided dialogues and reliable source citations, it enhances student learning and engagement. It also enables instructors to monitor student interactions and customize response granularity, ensuring accurate and adaptive support for different educational needs.

Currently in beta.

Check it out!

Fresh from MerryQuery!

Spring 2026

🎛️

GenAI Usage Scale

Replaced the Practice/Review binary with a 5-level GenAI assistance scale, giving instructors fine-grained control over AI help.

Spring 2026

📊

Analytics Dashboard

Launched the comprehensive Instructor Analytics Dashboard with engagement metrics, struggle detection, and topic performance charts.

Fall 2025

📚

Workspace Modes

Introduced Study and Practice workspace modes, allowing instructors to assign separate document sets for each learning context.

Discover MerryQuery's Features

Source Citation

MerryQuery generates responses with relevant course citations, allowing users to verify information and build trust.

GenAI Assistance Scale

Instructors can fine-tune how much help MerryQuery provides using a 5-level GenAI assistance scale — from Level 1 (concepts and pseudocode only) to Level 5 (full working solutions).

Guided Learning Dialogue

MerryQuery uses level-aware scaffolding to guide students without giving away answers. Each GenAI level enforces specific behaviors.

Instructor Analytics Dashboard

Instructors get a dashboard with engagement metrics, struggle detection, time-slot usage breakdowns, concept difficulty tracking, and topic performance visualization — plus tools to review student conversations and manage TA access.

Research-Driven Design

MerryQuery includes a built-in experiment framework supporting A/B testing with lab-section-based condition assignment.

Multimodal Data Processing

MerryQuery processes complex PDFs, using OCR and text mining to handle text, code, tables, and images.

Study Planning

Students have workspaces with a Study Plan page for guided study coaching and a Practice page for generating exam-style questions with tips and suggestions on their prompts.

Meet Our Team

Dr. Tiffany Barnes

Principal Investigator

Tiffany Barnes is a Distinguished Professor of Computer Science at NC State University, recognized for her contributions to AI in education and educational data mining. She has held various leadership roles in professional organizations and received an NSF CAREER Award for her innovative work in enhancing STEM learning environments.

Dr. Dongkuan (DK) Xu

Principal Investigator

Dongkuan (DK) Xu is an Assistant Professor at NC State, leading research in Artificial General Intelligence and generative AI systems, with numerous awards and publications in top AI conferences. His mission is to democratize AI applications and enhance productivity through collaborative generative intelligent agents.

Benyamin Tabarsi

Research Lead

Benyamin Tabarsi is a PhD student in Computer Science at North Carolina State University, specializing in CS education and AI. Passionate about using AI for social impact, he has worked on projects such as MerryQuery, SafeTalkCoach (an LLM-based medical assistant), and workshops training teachers on AI integration.