Mohmmad Alqallaf
Application Support Officer focused on banking technology, production support and reliable systems, with a broader background in software development, automation and research.
A little about me.
I am an Application Support Officer at Kuwait International Bank (KIB), working in IT application support with a focus on banking systems, cards, production support and incident investigation.
My work involves supporting business-critical applications, investigating technical issues, automation, monitoring, working with Linux and Oracle SQL, and coordinating with internal teams and vendors. Beyond operations, I have experience in software development and academic research.
Professional experience.
Application Support Officer
Kuwait International Bank (KIB)
Provide application support for core banking and card management systems, helping maintain operational stability and reliable service availability.
Monitor ATM, POS and online-channel transactions, investigate technical issues, analyze system behavior and support timely incident resolution.
Coordinate with internal teams, business users and external vendors to troubleshoot production issues, support system changes and maintain system performance.
Education.
Diploma, Information Technology
Credentials.
Verified professional certificates and credentials.
Google IT Support Professional Certificate
Professional credential covering core IT support, troubleshooting, networking, operating systems, system administration and security.
View credential ↗Research & publications.
Peer-reviewed work exploring practical uses of machine learning and deep learning, from identifying AI-generated images to forecasting temperature extremes across different climates.
Detecting Fake Images Generated by Artificial Intelligence Using Deep Learning Approach
This research investigates how reliably deep-learning models can distinguish AI-generated imagery from authentic images. Five CNN architectures were evaluated through a common training and validation pipeline, with DenseNet-121 and InceptionV3 reaching 98% accuracy and showing strong potential for content verification and multimedia forensics.
Machine Learning Models for Predicting Daily Temperature Extremes
This study compares Ridge Regression, SVR, Decision Tree, Random Forest, KNN and Neural Networks for predicting daily temperature extremes using data from Kuwait and South Korea. The results highlight how model effectiveness changes by climate and dataset characteristics while also examining the trade-off between predictive performance, interpretability and computational cost.
AI experiments & builds.
A hands-on lab for computer vision, deep learning and language-model work. Each entry opens into a dedicated page with the implementation, technical approach, repository results and an interactive visual explanation.
Image Colorization with U-Net
↗A PyTorch U-Net that predicts chrominance channels in Lab color space to generate plausible color from grayscale landscape images.
02
YOLOv8 · Detection
Pothole Detection
↗An end-to-end YOLOv8 object-detection pipeline covering dataset preparation, annotation conversion, training and inference.
03
YOLOv11 · Detection
Car Plate Detection
↗Vehicle license-plate detection using YOLOv11, including annotation conversion, training, evaluation and prediction visualization.
SpaceBot — Educational LLM
↗A compact educational chatbot fine-tuned on space-related questions using Microsoft Phi-2, LoRA and 4-bit quantization.