PhD researcher · University of Utah
Human-centered AI & data visualization
I'm a PhD student in Computer Science at the SCI Institute, University of Utah, advised by Prof. Valerio Pascucci.
I build and study tools that help people work with complex data: from AI agents for scientific exploration to accessible chart descriptions and visual interfaces for clinicians. My work brings together system building and studies with scientists, clinicians, and blind and low-vision readers.
Previously, I studied Computer Science and Engineering at BUET and worked on the data science team at IQVIA.
A few recent things
Service Serving as a Lead Student Volunteer at SC26, after volunteering at SC25.
Paper Chart-Supported or Model-Supplied? accepted at the AccessViz Workshop at IEEE VIS 2026, with Md Dilshadur Rahman.
Paper Our work on LLM-assisted animation of petascale data is published at ICBDA 2026.
Milestone Completed my MSc in Computer Science at the University of Utah.
Questions I'm excited about
I care about what happens between a person's question and a useful understanding of their data.
A multi-agent system that turns natural-language questions into analysis and visualization of large-scale scientific data.
Five specialized agents · 88 queries · Study with 10 domain scientists
Manuscript in preparationStudying how multimodal models distinguish what a chart supports from what a model adds, and what that means for accessibility.
102 visualizations · 3 models · 4 input conditions
Read the paper
Developing MagicInsight, a platform for annotating gigapixel pathology images, and studying how clinicians use visual tools.
Interaction capture & replay · User analytics · Clinician studies
Ongoing research & developmentResearch, shared
When an AI describes a chart, which claims come from the evidence, and which come from the model?
Helping scientists animate petabyte-scale climate data through conversational scripting on a commodity workstation.
Co-designing descriptions of complex set visualizations, evaluated with 11 blind and low-vision screen-reader users.
Connecting children in orphanages with medical consultation through trusted community intermediaries.
Accessible Text Generation for Complex Scientific Visualizations · AccessViz Workshop at IEEE VIS 2024
Enhancing Accessibility of Complex Plots · IEEE VIS 2024 poster
AccessViz Workshop ↗The path here
University of Utah · SCI Institute
Advisor: Valerio Pascucci · Expected 2029
University of Utah
Bangladesh University of Engineering and Technology
Thesis advisor: A. B. M. Alim Al Islam
SCI Institute, University of Utah
AI agents, scientific visualization & human-centered evaluation
IQVIA Asia Pacific
Data science & reporting systems
Free Pixel Games
Away from the screen
Dance has always been part of my life, from leading the BUET Dance Club to helping bring people together through the Bangladesh Student Association at Utah.
I also love exploring new places and cultures. Living in Utah has introduced me to hiking: a 16.7-mile Grand Canyon hike and the 14.2-mile Mt. Timpanogos summit are two favorites.
Let's connect
Interested in human-centered AI, accessible visualization, or scientific data? I'd love to hear from you.
