Experience

Graduate Research Assistant — University of Maryland, Baltimore County (UMBC)
Baltimore, MD, USA | Jun. 2026 – Jun. 2031

• Conducting a literature review to analyze and characterize mosquito movement.

• Developing a mosquito-movement digital-twin simulation using NVIDIA Omniverse Isaac Sim.

Graduate Research Assistant — Georgia Southern University (GSU)
Statesboro, GA, USA | Jan. 2026 – May 2026

• Developed FAR-TOD, an adaptive reconstruction-guided framework for tiny-object detection in UAV imagery, achieving 33.1 APvt while using 83% fewer tiles and 61.6% fewer GFLOPs than full-grid reconstruction.

Selected Projects

Adaptive Reconstruction-Guided Tiny-Object Detection for UAV Imagery

Related Publications

1. M. M. Alam, K. J. Tanha, M. T. Alam, S. Sakib, and M. G. Kim. FAR-TOD: Failure-Prior Adaptive Reconstruction Guided Tiny Object Detection in UAV Imagery. WACV, 2027. (Under review) [Paper]

2. M. M. Alam, K. J. Tanha, M. Golam, S. S. B. Rahman, E. A. Tuli, and T. Jun. Adaptive Reconstruction and Deformable Attention for UAV Small-Object Detection. IEEE ICMIC, 2026. [Paper Award] [Paper]

Graduate Research Assistant — Kumoh National Institute of Technology (KIT)
Gumi, South Korea | Mar. 2024 – Nov. 2025

Selected Projects

Real-Time Autonomous-Driving Hazard Analysis [GitHub]

• Developed a multimodal hazard-detection model integrating speed, vision, and audio.

• Developed a lightweight regression model for speed/audio-based driver-state detection and a vision-language model for visual detection.

Related Publications

1. M. M. Alam, K. J. Tanha, M. Kim, and T. Jun. Multi-modal Real-time Hazard Detection for Autonomous Driving Systems. KICS Fall Conference, 2025. [Paper] [GitHub]

AI for Digital Twin-Based Warehouse Logistics Optimization [GitHub]

• Developed digital-twin warehouse simulations in NVIDIA Omniverse Isaac Sim and FlexSim to generate synthetic data.

• Developed a DQN agent that reduced average picking time by 30.83% and average travel distance by 13.15%.

Related Publications

1. M. M. Alam, K. J. Tanha, M. T. Alam, S. Sakib, and M. G. Kim. AI-Driven Digital Twin Framework for Warehouse Logistics Automation Optimization. IEEE ICRA Workshop, 2026. [Paper]

2. M. M. Alam, M. Golam, M. R. Subhan, DS. Kim, and T. Jun. Federated Semi-Supervised Digital Twin for Enhanced Human-Machine Interaction in Industry 5.0. IEEE ICTC, 2024. [Paper]

3. M. M. Alam, R. A. N. Ugli, K. J. Tanha, and T. Jun. AI-Enhanced Digital Twin Systems for Warehouse Logistics Optimization: A Review of Challenges, Solutions, and Future Directions. ICT Express. [SCIE, IF: 4.6, Q1] [Paper]

4. K. J. Tanha, M. M. Alam, and T. Jun. Deep Q-Network Agent-Driven Dynamic Slotting Optimization in Digital Twin-Based Smart Warehouse Logistics. The International Journal of Advanced Manufacturing Technology. (Under review) [GitHub]

Daycare Face and Action Recognition on an Edge Device [GitHub]

• Developed a high-accuracy (98%) Transformer model for daycare face recognition using MTCNN facial embeddings.

• Achieved 97% action-recognition accuracy using a BiLSTM model with MoveNet pose embeddings.

• Constructed a Korean children's face-and-action recognition dataset.

Industrial Manufacturing-System Anomaly Detection [GitHub]

• Developed an attention-enhanced multi-CNN Transformer model achieving 98% accuracy.

• Used Stable Diffusion to generate synthetic defect images, mitigating data limitations and class imbalance.

• Improved baseline accuracy by 4.64% through optimized preprocessing, augmentation, and fine-tuning.

Related Publications

1. M. M. Alam, M. Golam, and T. Jun. Stable Diffusion-Augmented Industrial Anomaly Detection Using Attention-Enhanced Multi-CNN Transformer Fusion. IEEE Access. [SCIE, IF: 4.2, Q1] [Paper] [GitHub]

2. M. R. Subhan, M. M. Alam, and T. Jun. Enhancing Quality Control in MES Using Optimized Deep Learning. Measurement. (Submitted) [Pre-print]

Research Engineer — AIMS Lab, United International University
Dhaka, Bangladesh | Oct. 2023 – Dec. 2023

• Analyzed heart-failure, kidney-disease, and stroke data from rural areas of Bangladesh.

• Identified high-impact regions and investigated the factors associated with disease prevalence.

Software Engineer (Python) — Divine IT Ltd.
Dhaka, Bangladesh | Sep. 2022 – Sep. 2023

• Developed and implemented new modules and features for the PrismERP and PrismATS software suites.

• Improved system stability and performance by resolving bugs and optimizing parallel processing, algorithms, and database operations.

• Contributed to full-stack development for Enterprise Resource Planning (ERP) and Applicant Tracking System (ATS) products.