Antalya, Antalya, Türkiye
Software Engineering undergraduate with a strong research background in Data Science and Artificial Intelligence. I specialize in building end-to-end Machine Learning pipelines, from processing high-volume simulation data to deploying real-time Computer Vision models. My technical focus combines robust software engineering principles with advanced Deep Learning architectures.
• Spearheading the development of Artificial Intelligence models to analyze complex neutrino interactions within a high-energy physics context. • Conducting statistical analysis and data preprocessing on large-scale datasets to improve simulation accuracy.
• Led a funded research project focusing on signal-to-noise separation in Geant4 simulation data using Boosted Decision Trees (BDT). • Researched and applied Semantic Segmentation techniques (Mask R-CNN) to enhance topological feature extraction in detector outputs.
• Engineered end-to-end Machine Learning pipelines to classify Charged-Current (CC) and Neutral-Current (NC) interactions using Python, Pandas, and Scikit-learn. • Implemented and tuned gradient boosting models (XGBoost, LightGBM) achieving high accuracy in event separation tasks. • Performed rigorous Exploratory Data Analysis (EDA) and visualized complex data distributions using Matplotlib and Seaborn.
Analyzing CERN and Fermilab neutrino data under the discipline of Data Science and applying machine learning techniques to these data focuses on removing signal noise and correctly classifying interaction events using artificial intelligence-supported models.
• Developed a real-time Object Detection system distinguishing between UAVs and birds using YOLOv8 and PyTorch. • Curated a custom dataset involving data annotation, labeling, and augmentation to improve model robustness. • Deployed the trained model as a web application using Flask and OpenCV, enabling interactive inference for end-users.