Hello!

I'm Md Abdullahil Oaphy, an AI researcher working across LLMs, Multimodal Models, and Computer Vision to design human-centered, privacy-preserving intelligent systems.

Get in touch abdullahil.oaphy@gmail.com , oaphyapran365@gmail.com

Background

I am currently a Graduate Research Assistant (GRA) at Kennesaw State University, working under the supervision of Dr. Honghui Xu . My research centers on advancing Large Language Models (LLMs), Multimodal AI, and privacy-preserving machine learning. I develop enhanced multimodal LLMs through LoRA fine-tuning and Dual-Differential Privacy, improving their robustness and reducing harmful or inaccurate generations. I also design lightweight YOLO-based UAV vision models optimized for efficient inference on edge hardware.

Prior to KSU, I worked as a Junior Research Engineer at the DIU NLP & ML Research Lab in Bangladesh, where I built deep learning models for MRI brain tumor classification and segmentation, contributing to improved diagnostic support systems. I also applied transfer learning to rare-bird species detection, supporting biodiversity research through scalable and reliable computer vision solutions.

Earlier, I served as a Software Developer at Masleap Plc., developing complete full-stack applications using ReactJS, Django, and MongoDB. I integrated Intel OpenVINO for real-time video and image analytics, enhancing system responsiveness and improving data-driven decision capabilities for business operations.

My journey into AI began as a Machine Learning Engineer Intern at Nazpev Inc. in Japan, where I worked on time-series forecasting for retail sales data to support more accurate and reliable demand planning workflows.

Across all my roles, I have enjoyed working at the intersection of AI research, engineering, and real-world impact. I am passionate about building intelligent systems that are technically robust, privacy-aware, and aligned with human needs. I continually strive to bring meaningful, ethical, and scalable AI solutions to life.

Education
Jan 2025 – May 2026 (Expected)
Kennesaw State University, Marietta, GA, USA
Daffodil International University, Dhaka, Bangladesh
GPA: 3.69 / 4.00
Skills
Languages
  • Python
  • C++
  • C
  • JavaScript
  • SQL
AI/ML Frameworks & Tools
  • PyTorch
  • TensorFlow
  • Scikit-learn
  • NumPy
  • Pandas
  • OpenCV
  • Hugging Face Transformers
  • LoRA Fine-tuning
  • LangChain
Data Analytics & Visualization
  • Tableau
  • Power BI
  • Excel (PivotTables, Power Query)
  • ETL Pipelines
  • Matplotlib
  • Seaborn
Web & API Development
  • FastAPI
  • Flask
  • Django
  • REST API Development
MLOps, Cloud & Deployment
  • Linux (Ubuntu)
  • AWS
  • Docker
  • Git / GitHub
  • Weights & Biases
  • MLflow
  • ONNX Runtime
  • TensorRT
Work Experiences
Jan 2025 – Present
Kennesaw State University, Marietta, GA
Jan 2024 – Oct 2024
DIU NLP & ML Research Lab, Dhaka, Bangladesh
Jan 2022 – Jul 2022
Masleap Plc., Dhaka, Bangladesh
Feb 2021 – Mar 2021
Nazpev Inc., Ichikawa City, Chiba, Japan
Projects

Developed a YOLO-based detection framework for UAV-based post-disaster building damage assessment, integrating DP-SGD and structured pruning to jointly optimize accuracy, privacy, and deployment efficiency on edge devices.

PyTorch YOLO Differential Privacy Structured Pruning UAV Imagery

Trained DeepLabV3+ (ResNet-101) for road-scene segmentation across fog, night, rain, and snow conditions, using weather-aware augmentations and BN adaptation to improve robustness and stability for autonomous driving perception.

PyTorch DeepLabV3+ ACDC Dataset OpenCV Semantic Segmentation

Built a multimodal Retrieval-Augmented Generation pipeline that combines chest X-rays and clinical reports using Qwen2-VL, CLIP, BM25, FAISS, and a cross-encoder reranker, with a FastAPI backend and Gradio UI for interactive medical question answering.

Qwen2-VL CLIP BM25 FAISS FastAPI Gradio

Implemented Dual-Differential Privacy with two-stage noise injection (embedding-level and LoRA parameter-level) for multimodal LLMs, evaluating accuracy, hallucination rate, and privacy budget to study privacy–utility trade-offs in image–text alignment.

MiniGPT-4 Vicuna LLaMA2 LoRA Differential Privacy Multimodal LLMs

Developed a lightweight, privacy-preserving multimodal framework that fuses UAV imagery and social-media text for disaster event classification. Combines ResNet50 visual features and BiLSTM + GloVe text embeddings in a late-fusion architecture, with DP-SGD for ε-differential privacy and structured neuron pruning for efficient edge deployment on UAVs and field devices.

ResNet50 BiLSTM GloVe Multimodal Fusion DP-SGD Differential Privacy Structured Pruning Edge Deployment

Benchmarked ARIMA, Prophet, LSTM, and XGBoost on the UCI Online Retail II dataset to build a robust revenue forecasting pipeline, and designed a pricing optimization engine with demand elasticity modeling to support data-driven pricing and retention strategies.

Python ARIMA Prophet LSTM XGBoost Time-Series Forecasting
Research Publications
Certification
Computer Vision Basics
Coursera
Perform Real-Time Object Detection with YOLOv3
Coursera
Capstone: Retrieving, Processing, and Visualizing Data with Python
Coursera
Python Data Structures
Coursera
Programming for Everybody (Getting Started with Python)
Coursera
AWS Fundamentals
Coursera
AWS Fundamentals: Migrating to the Cloud
Coursera
Achievement
CCSE Graduate Student Travel Award
2025
Kennesaw State University — Research travel funding support.
Research Grant
2021
Division of Research, Daffodil International University, Bangladesh
DIU merit-based Scholarship on tuition fee
2017–2021
Daffodil International University
Talent Pool Scholarship
2008–2012
Bangladesh Government
Community Service and Volunteering Activities
Teacher & Mentor
04 Months
'Play with Python' Crash Course
Executive Member
01 Year
DIU NLP & ML Lab