Anindya Das Antar

“If not now, when? If not you, who?”
― Hillel the Elder

---

I’m currently on the tenure-track faculty job market (at the intersection of technical HCI and AI, with an emphasis on system design, information visualization, human-AI alignment, and AI knowledge elicitation). Feel free to reach out at aantar[at]nd[dot]edu or adantar[at]umich[dot]edu if you are hiring or would like to collaborate.

I am currently a Lucy Family for Data and Society Postdoctoral Research Fellow at the University of Notre Dame. I am also affiliated with the Human-Centered Responsible AI Lab and the Department of Computer Science and Engineering, working with Toby Lee. My current work explores how interactive AI agents can support complex human decision-making and enhance visibility into inequality, as well as facilitate coordination toward fairness across high-stakes domains, including gig-work platforms and student well-being.

My PhD research focused on developing human-centered behavior models and explaining their decision-making processes to both model engineers and end-users (e.g., domain experts, policymakers, and consumers) who may not have formal AI expertise. I develop human-centered interactive tools, akin to “Photoshop for AI” toolkits, designed to support users across various stages of the AI model lifecycle (e.g., pre-training, development, and post-deployment). These toolkits enable users to first conduct “what-if” scenario exploration to explore data- and algorithm-centric missing (latent) domain knowledge in ML models. They can then apply “guardrail” tools to edit the model’s input-output relationships and incorporate missing domain knowledge through Bayesian inference-driven hierarchical meta-models.

I also investigate how users can interactively audit pre-packaged large language models (LLMs) to critically reflect on model behavior and enhance their AI literacy. I am also developing interactive tools to support “user–LLM” alignment. The goal is to select effective guardrails and fine-tuning mechanisms that help shape LLM outputs to align with expert users’ expectations, domain values, empirical findings, societal goals, policy standards, industry requirements, copyright constraints, and ethical norms.

I completed my Ph.D. in Computer Science and Engineering at the University of Michigan in August 2025. From the same department and institution, I received my masters on my way to Ph.D. I received my bachelor’s in Electrical and Electronic Engineering from the University of Dhaka. Before beginning my Ph.D., I worked as a visiting researcher at Osaka University. Before beginning my Ph.D., I worked as a visiting researcher at Osaka University.

Research Interests

• Applied Machine Learning
• Human Behavior Modeling
• Human-AI Interaction (HAI)
• Human-centered {Interactive} Explainable AI (HCXAI)
• Responsible AI with a focus on human-AI alignment

Recent Updates

• Oct 2025: Joined the University of Notre Dame as a “Lucy Family Institute for Data and Society” Postdoctoral Research Fellow.
• Oct 2025: Attended the AIES 2025 conference in Madrid
• August 2025: Defended my Ph.D. (Yay!)
• Got interviewed by Michigan Daily on “UMich introductory computer science courses face possible shift with CS open letter”— by Kayla Lugo

Older Updates

• I worked as a student volunteer for 7th Summer School on Computational Interaction, University of Michigan. Weblink
• Presented at the Ubicomp 2023 conference in MexicoI worked as a co-mentor and teaching assistant for Machine Learning Research group, Big Data Summer Institute 2021, School of Public Health, University of Michigan. Weblink
• Our Research group was featured in Michigan News Center for research contributions employing AI to help address COVID-19 challenges.
  Link: Understanding the Effect of Human Behaviors on Social Distancing,
  How predictive modeling could help us reopen more safely
  Video: https://youtu.be/KzfIW6-hjj4
• Received Weinberg Institute for Cognitive Science Fellowship by University of Michigan
• Started PhD in the Department of Computer Science and Engineering, Computational HCI Lab, University of Michigan, Ann Arbor under Prof. Nikola Banovic.
• A paper titled “Action recognition using Kinematics Posture Feature on 3D skeleton joint locations” accepted in Pattern recognition Letters Journal 2021. (This work was done in Yagi Lab, Osaka University)

Portfolio: Anindya Das Antar