Muhammed Zeki Kurt

I'm a recent Computer Engineering graduate from Eskişehir Technical University building practical data science and artificial intelligence projects in Python.

I developed a personalized glucose prediction model using hybrid LSTM and CNN-LSTM architectures, achieving 93% accuracy through data preprocessing, feature engineering, training, and hyperparameter optimization.

For SentiFuzzy-TR, I built a Turkish sentiment analysis engine combining BERTürk, Zeyrek morphological stemming, PyTorch, Hugging Face Transformers, and Fuzzy Logic. I also deployed its dual-engine inference system as a FastAPI RESTful microservice and containerized it with Docker.

My work also includes Google Play Store and IMDb analyses, retinal-image CNN classification, and real-time social-media sentiment tracking. During my IT internship at Karayolları Genel Müdürlüğü, I supported IT infrastructure installation, hardware maintenance, configuration, and technical support.