Ayesha Siddiqa

PROFESSIONAL PROFILE I am a motivated Software Engineering undergraduate at NUST (CGPA 3.88) with a strong interest in machine learning and real- world problem solving. i have hands-on research experience in remote sensing, satellite imagery analysis, and transformer- based models, including work on land-use change detection, deforestation monitoring, and vision-language models. Comfortable working across the full ML pipeline data collection, preprocessing, model training, evaluation, and deployment using Python and modern ML frameworks. I have been recognized for consistent academic excellence and leadership, with the ability to translate theoretical concepts into reliable, well-engineered systems. Actively seeking roles where strong fundamentals in software engineering and applied machine learning can be used to solve complex, real-world problems. EDUCATION Bachelor of Software Engineering SEECS , Islamabad , 3.88 (2026) INTERNSHIP EXPERIENCE Machine Vision and Intelligent Systems Lab, SEECS, NUST 11-Jun-2025 - 31-Aug-2025 Conducted research on remote sensing and satellite image analysis to monitor deforestation and urban expansion. Created custom bi-temporal datasets using Google Earth Engine for 20 global regions to track urban expansion and land use changes.Trained and optimized machine learning models and transformer architectures (e.g., BIT, ScratchFormer) for land cover change detection Machine Vision and Intelligent Systems Lab, SEECS, NUST 01-Sep-2024 - 01-Aug-2025 Working on generating detailed textual descriptions of satellite imagery using pre-trained Vision-Language Models (VLMs).In parallel, worked with point cloud data to analyze spatial structure and elevation-based features, supporting tasks such as urban expansion analysis and scene understanding. This work emphasizes multimodal learning, geospatial data pipelines, and the practical challenges of aligning visual, spatial, and textual representations for real-world remote sensing applications. FINAL YEAR PROJECT Gaze-Guided Explainable AI for EEG Brain Disorder Classification This project is a human-aligned AI framework that integrates neurologist eye-tracking data with EEG signals to make deep learning based clinical decisions transparent, verifiable, and clinically meaningful. The project captures where experts visually focus during EEG interpretation and synchronizes this gaze information with EEG epochs to create multimodal datasets combining electrophysiology, attention maps, and diagnostic labels. By training models to align their internal attention and explanations with expert gaze patterns, the system addresses key limitations of black-box EEG classifiers , lowering cognitive load and enabling clinicians to validate whether predictions are based on medically relevant waveform features rather than spurious correlations. The outcome is an interpretable, trust-worthy AI system that bridges human expertise and machine intelligence, accelerating EEG analysis while preserving clinical rigor and accountability. TECHNICAL EXPERTISE Machine Learning

I am a motivated Software Engineering undergraduate at NUST (CGPA 3.88) with a strong interest in machine learning and real- world problem solving. i have hands-on research experience in remote sensing, satellite imagery analysis, and transformer- based models, including work on land-use change detection, deforestation monitoring, and vision-language models. Comfortable working across the full ML pipeline data collection, preprocessing, model training, evaluation, and deployment using Python and modern ML frameworks. I have been recognized for consistent academic excellence and leadership, with the ability to translate theoretical concepts into reliable, well-engineered systems. Actively seeking roles where strong fundamentals in software engineering and applied machine learning can be used to solve complex, real-world problems.