Contact
Education
University of Utah, Salt Lake City, Utah
Ph.D. in Computer Science
Kahlert School of Computing | August 2023 – Present
CGPA: 3.978 / 4.0 (current)
Advisor: Dr. El Kindi Rezig
Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh
B.S. in Computer Science
April 2017 – April 2022
Advisor: Dr. Muhammad Abdullah Adnan
Research Interests
I focus on data preparation in general and data discovery in particular for data lakes. I built SemDisc, the first end-to-end semantic join discovery system using a query-by-example interface. I also explored how discovered relationships can recommend meaningful ways to organize and sort data. Additionally, I contributed to Buckaroo, a visual data wrangling system that enables users to interactively clean and repair data anomalies through direct manipulation of visualizations. My interests include:
Data Systems • Data Lakes • Data Discovery & Integration • Data Wrangling • Data Cleaning • AI for Data Management
Publications
Publications since starting Ph.D. (2023-Present, ~2.3 years)
- SIGMOD'26: , El Kindi Rezig. "Qualitative Join Discovery in Data Lakes using Examples." Accepted at ACM SIGMOD International Conference on Management of Data (SIGMOD'26), 2026. [PDF] a system for discovering hybrid join paths (combining semantic and equi-joins) in data lakes using query-by-example, supporting hidden tables and semantic tuple matching
- CIDR'26: El Kindi Rezig, , Nicolas Baret, Ricardo Mayerhofer, Andrew McNutt, Paul Rosen. "Towards Scalable Visual Data Wrangling via Direct Manipulation." Accepted at CIDR 2026. [PDF] a visual data wrangling system that enables users to clean and repair data anomalies through direct manipulation of interactive visualizations
- VLDB'25 (Demo): Akash Khatri, , El Kindi Rezig. "Sort it Like You Mean It: Discovering Semantically Interesting Attribute Augmentations to Sort Tables." Accepted at VLDB 2025 (Demo Track). [PDF] Recommends semantically meaningful ways to sort tables by automatically discovering and augmenting attributes from data lakes using LLMs.
Undergraduate research
- IEEE FG'24: Iftekhar E Mahbub Zeeon, , Muhammad Abdullah Adnan. "BTVSL: A Novel Sentence-Level Annotated Dataset for Bangla Sign Language Translation." Accepted at IEEE FG 2024. Introduces the first large-scale sentence-level dataset for Bangla Sign Language translation, derived from 60 hours of YouTube news content with professional signers. [PDF] [Link]
- Neurocomputing'22: Md. Ashraful Islam, , Sarkar Snigdha Sarathi Das, Mohammed Eunus Ali. "A survey on deep learning based Point-of-Interest (POI) recommendations." Accepted at Neurocomputing (Journal), 2022. [PDF] [Link] Categorizes deep learning approaches for POI recommendation systems in location-based social networks.
Research Experience
University of Utah, Kahlert School of Computing, Salt Lake City, UT
Graduate Research Assistant, August 2023 – Present
Advisor: Dr. El Kindi Rezig
- Developed algorithms for qualitative join discovery in data lakes using example-based queries, enabling efficient dataset integration across heterogeneous tabular data
- Built systems for semantic attribute augmentation and table sorting, improving data discovery workflows for analysts working with complex datasets
- Implemented scalable data wrangling systems with direct manipulation interfaces, handling data transformations efficiently
Bangladesh University of Engineering and Technology, CSE, Dhaka, Bangladesh
Research Assistant, July 2022 – June 2023
Advisor: Dr. Muhammad Abdullah Adnan
- Developed machine learning pipelines for processing and analyzing large-scale video datasets for sign language translation
- Built data collection and annotation systems for creating structured datasets, handling data cleaning
Additional Experience
Everforth, Tokyo, Japan (Remote)
Frontend Developer, April 2022 – June 2023
- Built scalable web applications using Vue.js and CakePHP
Technical Skills
Selected Projects
- Badhan Blood Donation Management System: Designed and implemented a full-stack blood donation platform with MongoDB backend, serving users across BUET campus with real-time donor matching and request management [GitHub]
- CNN-Based Object Detection: Developed deep learning models for real-time object detection using PyTorch[GitHub]
- Automated Robotic Arm: Built computer vision and control systems for robotic manipulation tasks using MATLAB, integrating sensor data processing and motion planning algorithms [GitHub]
- Comparative Analysis of AI Agents for Othello: Compared 12 Othello AI agents, from heuristic search baselines to reinforcement learning approaches, using a round-robin tournament to analyze their relative performance [GitHub]