A Computer Engineer with research experience in machine learning for medical diagnostics and IoT-based environmental monitoring — building solutions that make a real difference.
Download CV
Choosing computer engineering was never just about software — it was about impact. From developing a deep learning model for ocular cancer detection to building IoT sensor systems for environmental safety, the question driving my work has always been the same: “Can this knowledge make a real difference?”
Today I channel that drive into machine learning research and smart systems — working on projects like AI-powered anemia monitoring and crowdfunding data analysis. Every project is a step toward technology that genuinely improves people's lives.
Semafor Technology
Trained and developed a machine learning model for electricity outage prediction in the energy domain, focusing on forecasting power interruptions based on historical and operational data.
Dokuz Eylul University Research Lab
Contributed to a research project utilizing deep learning for medical diagnostics. Developed a machine learning model for ocular cancer detection via image analysis, achieving 95% accuracy. Collaborated with faculty to enhance model performance for early cancer detection.
Dokuz Eylul University Research Lab
Designed a smart sensor system with Arduino, integrated into clothing for environmental monitoring. Developed the system to detect toxic gases and smoke, providing real-time alerts. Implemented cloud storage for environmental data analysis and long-term storage.
Computer Engineering, M.Sc.
Computer Engineering, B.Sc.
Mathematics & Physics
Generative AI Bootcamp
Completed Akbank's Generative AI Bootcamp — a hands-on program covering large language models, prompt engineering, retrieval-augmented generation (RAG), and building AI-powered applications with modern generative AI tools.
Problem
Early detection and monitoring of anemia requires accessible and intelligent tools for patients.
Solution & Impact
An AI-powered app that analyzes medical reports (PDFs) and provides insights to assist in early anemia detection and monitoring.
Problem
Finding relevant recipes based on available ingredients is time-consuming and often yields irrelevant results.
Solution & Impact
Implemented a RAG architecture combining TF-IDF/Cosine Similarity for retrieval with Google Gemini (gemini-2.5-flash) for generation. Interactive chatbot UI built with Streamlit over a 13k-recipe dataset.
Problem
Identifying the key factors that lead to success in Turkish crowdfunding campaigns.
Solution & Impact
Machine learning models and R language to analyze campaign data and predict the factors that drive startup success.
Languages & Frameworks
Methods & Concepts
I'm always open to new projects, collaborations, and conversations. Reach out via the form or connect on social platforms.