cv

Doctor of Philosophy, Computer Science

University of Illinois Chicago

• Advisors: G. Elisabeta (Liz) Marai and Fabio Miranda

Master of Science, Computer Science

University of Illinois Chicago

• Advisors: G. Elisabeta (Liz) Marai and Fabio Miranda

Bachelor of Science, Computer Science & Engineering

Military Institute of Science and Tehcnology, Bangladesh

• Advisors: Md. Mahbubur Rahman

Graduate Research Assistant

University of Illinois Chicago

• Design and build human-centered visual analytics systems for complex, heterogeneous, multivariate, and spatiotemporal data, enabling interpretable, uncertainty-aware, and actionable analysis across real-world domains.
• Develop declarative frameworks and scalable pipelines (e.g., StreetWeave) that lower authoring barriers and support reproducible, multi-resolution analysis for spatial and network-based data.
• Investigate large language models and agentic AI for visual analytics, studying how LLMs and multi-agent systems author and decompose analytical workflows, and how interaction with visual analytics improves the quality and interpretability of agent-generated insights through human-in-the-loop interfaces.
• Develop machine learning prediction models and visual analytics systems for complex clinical and urban domains, including head and neck cancer outcome analysis and street-level urban analytics, in collaboration with domain experts and policy stakeholders.
• Apply human factors and UX research methods, including controlled user studies with 78 participants, workshops, semi-structured interviews, thematic coding, and mixed-methods analysis, to understand user needs, evaluate system usability, and inform evidence-based design decisions.

Graduate Teaching Assistant

University of Illinois Chicago

• Courses: Visual Data Science, Data structure, Computer design & Virtual, Augmented & Mixed Reality, Game Design, and Computer Design.
• Conducted office hours, graded lab presentations, assignments and projects.

Data Visualization and Machine Learning Intern

Genentech

• Collaborated with the gCS team (5 computational biologists, 3 ML engineers) to build a modular, browser-based VA tool using ONNX Runtime WebAssembly for sequence-based gene expression models, supporting real-time inference, gene-expression and single-cell analysis, and LLM-assisted in-silico mutagenesis, attention analysis, and multi-sequence and cell-type comparison.
• Translated client needs into features and milestones, delivered an MVP, and iterated via usability studies; enabled no-code hypothesis testing, reduced time to insight by 6\%, and removed backend compute; piloted by 15 researchers.

President

Girls Who Code College Loop, University of Illinois Chicago

• Improved retention in computer science and related disciplines.
• Helped in building community and developing a supportive network of female peers and connecting Girls Who Code alumni on campus.

Lecturer

Department of Computer Science and Engineering, Eastern University

• Supervised undergraduate courses e.g., Web development, Digital logic Design, C/C++, Java and Matlab programming classes and labs.
• Worked as assistant course coordinator and student advisor.

Dysphagia Risk Stratification in Head and Neck Cancer

• Led development of a two-stage machine learning framework combining XGBoost with ElasticNet logistic regression to identify patients at risk for moderate-to-severe swallowing impairment.
• Built leakage-resistant 5-fold cross-validation and 50-seed evaluation pipelines, achieving 0.885 AUC and 0.783 balanced accuracy with SHAP-based model interpretation.

COMPASS: AI-Assisted Clinical Visual Analytics for Patient Progression

• Developed an AI-assisted healthcare analytics system that transforms EMR-derived notes into reusable clinical event timelines, evidence trends, and discipline-specific Q&A components for multidisciplinary patient review.
• Collaborated with 6 clinical teams to scope requirements, validate AI-supported outputs, and translate domain expertise into progression indicators for fall and NICU cases.

Conversational AI-Driven Glyph Visualization for Enhanced Multidimensional Data Communication

• Leveraged LLMs (GPT-4 Vision, ChatGPT) to extract glyph elements from hand-drawn sketches and semantically map them to data attributes via conversational text prompts.
• Developing an LLM-powered visual analytics system that guides novice users to create expert-level glyph visualizations, reducing design complexity and enhancing accessibility.

Street and Pedestrian Network Visualization: Design Framework and Declarative Grammar Development

• Reviewed 45+ studies on street-overlaid visualizations to identify design challenges and propose a design space for multivariate spatial network visualization.
• Developed StreetWeave, a declarative grammar and scalable data pipeline to simplify the creation of street network visualizations, reducing technical barriers for domain experts by eliminating low-level coding. Enabled reproducible, multi-resolution visualizations integrating thematic and physical data, adopted in real-world case studies for cross-domain spatial analysis and decision support.
• Implemented a weather-aware routing algorithm using time-varying forecasts and GNN-based edge impact estimation to compute routes that reduce adverse weather

Visualization of Probability Distributions of Geographical Data

• Adapted non-spatial probability encodings to geographic data, solving challenges in quantifying uncertainty.
• Designed and conducted a controlled user study with 78 participants to evaluate visualization accuracy, completion time, confidence, and usability.

Health Disparities in the Head and Neck Cancer Patient Population

• Harmonized two large, multi-institutional HNC datasets (UI Health & MD Anderson) to support comparative analysis of health disparities.
• Developed visual analysis tool for HNC data using machine learning approach, enabling oncologists to analyze demographics, disease characteristics, and treatments and outcomes differences across cohorts.

Analyzing County-level COVID-19 Symptoms Pattern and Risk-factors

• Analyzed county-level COVID-19 symptoms pattern and risk-factors correlated with hospitalization rate, ICU treatments and morbidity rate using data mining and machine learning approaches.
• Analyzed big data consists of 300 million rows.

Visual Analysis of COVID-19 Forecast Ensemble Models

• Built COVID-19 EnsembleVis, an open-source, browser-based system for evaluating pandemic forecast models at county-level granularity, exposing spatial disparities often missed at coarser scales.
• Enabled comparison of ensemble vs. individual models through spatial-temporal visualizations, helping researchers quantify uncertainty and assess prediction reliability across geographies.

Visualizing Similarities and Differences between Head and Neck cancer patients

• Developed 3D visualization system for head and neck cancer patients.
• Visualized patients similarities and dissimilarities in terms of dose prescription, dose distribution, tumor size, and presence of lymph node.

Android Based Advanced Attendance System usingWireless Network and Bio-metric Fingerprint Authentication

• Designed and developed a smart attendance system using a custom Android app integrating biometric fingerprint authentication and Wi-Fi connectivity to verify both identity and physical presence.
• Implemented session-based login and real-time tracking to prevent proxy attendance and ensure continuous presence monitoring.

StreetWeave: A Declarative Grammar for the Visualization of Multivariate Data for Spatial Networks

• Paper presented at IEEE VIS in Vienna, 2025.

Unveiling Complexities: Visualization Techniques for Heterogenous Multivariate Data Analysis

• Ph.D. Proposal Defense, 2024.

Interactive Exploration of Inferences from Genomic Sequence-Based Machine Learning Models

• Presented my internship work at Genentech, summer 2024.

A Tale of Two Centers: Visual Exploration of Health Disparities in Cancer Care

• Paper presented at IEEE PacificVis and at Visualization and Data Analytics Research Center at NYU, 2022.

Approaches for Uncertainty Visualization of Temporal Data

• Ph.D. Qualifier Exam, 2021.

COVID-19 EnsembleVis: Visual Analysis of County-level Ensemble Forecast Models

• Paper presented at IEEE Workshop on Visual Analytics in Healthcare (VAHC), IEEE VIS virtually and in person at UIC, EVL satellite event for the conference.

Android based advanced attendance vigilance system using wireless network with fusion of biometric fingerprint authentication

• Paper presented at International Conference on Advanced Communication Technology (IEEE ICACT), 2018, South Korea.

GiveMed: A webportal for medicine distribution among poverty-stricken people

• Paper presented at IEEE Region 10 Humanitarian Technology Conference (R10-HTC), 2017.

GHC Scholarship

• Grace Hopper Celebration (GHC) Scholarship for attending GHC conference 2023

Dean’s List

• MIST Dean’s List of honor for excellent result – in the years 2017, 2016, 2015, and 2014

Merit Scholarship

• University Merit Scholarship for excellent result – in the years 2017, 2016, 2015, and 2014

Student Volunteer

• Provide essential support to the IEEE VIS Conference: 2022, 2025 (in-person), 2021 (remotely).

Conference Reviewer

• IEEE VIS: 2022, 2023, 2024 - Reviewed both full and short papers.
• EuroVIS: 2023, 2024 - Reviewed full papers.

IEEE VIS Satelite Event

• Organized and managed IEEE VIS Satelite Event at the Electronic Visualization Laboratory, 2021.

UIC Open House

• Organized and managed public visits at the Electronic Visualization Laboratory, 2019.