Shi Ding

Human-Centered AI · AI Agents · AI Evaluation
Georgia Institute of Technology

About

I am a Ph.D. Candidate at the Georgia Institute of Technology, advised by Regents Professor Brian Magerko in the Expressive Machinery Lab.

My research lies at the intersection of human-centered AI, agentic AI, and AI evaluation. My goal is to make AI systems more reliable, transparent, and aligned with human needs.

At Georgia Tech, I design, develop, and evaluate human-facing AI systems through real-world deployments, including the NSF-funded EarSketch project , where I investigate human-AI interaction, LLM-powered learning, and accessibility in educational contexts.

Selected Projects

Human-Centered AI Interactive AI RAG

EarSketch AI Assistant

A context-aware AI assistant for music programming that combines conversation history, curriculum resources, code state, and platform tools to support learners during open-ended coding tasks. I worked on system design, deployment, and multi-turn evaluation.

Visit EarSketch

Accessibility Co-Design Qualitative Research

Accessible EarSketch

A qualitative co-design research project with blind and visually impaired high school learners and educators, identifying requirements for accessible and context-aware future AI support in creative coding.

🏆 ACM IDC 2026 Best Paper Honorable Mention

Paper

AI Infrastructure Human-Centered AI Agentic AI

Deeproject AI Evaluation Platform

A human-centered AI platform combining evaluation methods with scalable infrastructure for assessing interactive AI systems.

Designed to support more reliable AI development and continuous improvement.

Co-founder 🦄· Georgia Tech CREATE-X QI Fellow · 2025–2026

News

Jul 2026 Received the 🏆 Honorable Mention for the Best WIP Award at ACM IDC 2026

Jun 2026 First-author paper accepted to ISLS 2026 and presented at UC Irvine (CS Track)

Jun 2026 Successfully passed Ph.D. proposal defense

Jun 2026 First-author paper accepted to the ACL 2026 EvalEval Workshop

Selected Publications

ACL 2026

Evaluating Multi-turn Human-AI Interaction

Shi Ding, Sijian Tan
EvalEval Workshop at the Annual Meeting of the Association for Computational Linguistics (ACL), 2026
NeurIPS 2025

Rethinking AI Evaluation through TEACH-AI: A Human-Centered Benchmark and Toolkit for Evaluating AI Assistants in Education

Shi Ding, Brian Magerko
Socially Responsible and Trustworthy Foundation Models Workshop at NeurIPS, 2025
AIED 2025

Considering Large Language Model Integration in Expressive Computer Science Learning Environments for Blind and Visually Impaired Learners through Co-design

Shi Ding, Jason Brent Smith, Brian Magerko
Proceedings of the International Conference on Artificial Intelligence in Education (AIED), 2025
SIGCSE 2026

Bridging Responsible AI and AI Literacy: The TEACH-RAI Framework and Toolkit for Education, Design, and Research

Shi Ding, Brian Magerko
Proceedings of the 57th ACM Technical Symposium on Computer Science Education (SIGCSE TS), 2026
IDC 2026 Award

Human–AI Interaction for Accessible CS Learning: Co-Designing AI with Blind and Visually Impaired Learners

Shi Ding, Jason Brent Smith, Kevin Gautier, Stephen Garrett, Brian Magerko, Jason Freeman, Bryan Pardo, Stephanie Ludi, Taneisha Lee, Tom McKlin
Proceedings of the ACM Interaction Design and Children Conference (IDC), 2026

🏆 Honorable Mention for the Best WIP Award

Awards & Service

🏆 Honorable Mention for the Best WIP Award, ACM IDC, 2026
AI Child Safety Lab Fellowship Recipient, 2026
CRIDC Research Conference Award Recipient, Georgia Institute of Technology, 2025
Invited Talks: Harvard University, 2026; Georgia Institute of Technology, 2025
Area Chair, DIS 2026 Works-in-Progress Track
Session Chair, ACL 2026 NLG, RAG, and Efficiency Track
Reviewer for CHI, DIS, FAccT, AMIA, C&C, NeurIPS, SIGCSE TS, and CogSci, 2025–2026
Co-founder, Deeproject AI; Quadrant-i Startup Grant Recipient, Georgia Institute of Technology, 2025–2026

Contact

sding84@gatech.edu