Ampersand Summary - The Current Chapter Captured in Nazaré, April 2026
Father & Husband
Technology & Research
Professional Practice & Academic Curiosity
Data & Earthquakes
Istanbul & Barcelona
Me, Myself & I
About Me
- Welcome to my personal webpage! I'm Utku Unal, a tech enthusiast passionate about Technology, Geophysics, Big Data, AI, and creating innovative solutions.
- I am a proud father of a teenage boy.
- I am a technology executive and researcher working at the intersection of artificial intelligence, geophysics and large-scale computing. After more than 30 years in enterprise technology, I returned to academia to study how machine learning can improve seismic phase detection and earthquake analysis.
- I hold a degree in Computer Engineering from Bogazici University and I have a master’s degree in geophysics from Bogazici University's KOERI (Kandilli Observatory and Earthquake Research Institute). I pursued Master on Artificial Intelligence in UPC (Universitat Politècnica de Catalunya) and my master thesis transitioned into PhD research on machine learning in Geophysics (seismic phase picking) at Canakkale Onsekiz Mart University. I am currently pursuing a PhD in Canakkale Onsekiz Mart University, School of Graduate Studies, Geophysical Engineering. My area of research is using deep learning for earthquake parameter analysis with seismological big data. Feel free to explore the links below to learn more about me and my work (Orcid: 0009-0008-6124-4013): https://orcid.org/0009-0008-6124-4013
- You can read my short stories in Turkish at Derkenar E-Magazine: https://derkenar.com/yazar/utku-unal
My Research
- Unal, U., & Bekler, T. (2025). Detection of P and S wave phases by machine learning using north-western Türkiye local seismic network data. Pure and Applied Geophysics, 182, 1381–1395. https://doi.org/10.1007/s00024-024-03636-4
- Unal U., Bekler T., Bekler F. N. Makine Öğrenmesi Algoritmalarının Türkiye Depremlerinde P ve S Fazlarının Belirlenmesindeki Etkinliği. Türk Deprem Araştırma Dergisi (2025). https://doi.org/10.46464/tdad.1597618
- Unal, U., Bekler, T. TroyNet: A Region-Specific Deep Learning Model for Seismic Phase Picking in Noisy Networks of Northwestern Türkiye. Pure Appl. Geophys. 183, 1621–1645 (2026). https://doi.org/10.1007/s00024-026-03938-9
TroyNet is a deep learning model developed for automatic P- and S-wave arrival picking in noisy seismic recordings from northwestern Türkiye. In a controlled comparison with a retrained PhaseNet baseline, it achieved higher recall for both phases while maintaining comparable precision.
Download the accepted manuscript (PDF)
This is the peer-reviewed accepted manuscript, prior to publisher copy-editing and typesetting. For citation, please refer to the published article.
My Professional Journey
- My career in technology began in 1992, when I worked as a programmer and computer teacher while developing a complete school management system. Since then, I have built more than three decades of experience across software development, IT consulting, cybersecurity, managed services, cloud computing, and technology leadership.
- I spent over 16 years at Hewlett-Packard, progressing from technical consulting and project management to senior leadership roles. I later held regional and executive positions at Huawei, Fujitsu, and Oracle, leading large multidisciplinary teams, complex technology programmes, and multimillion-euro operations across Türkiye, the Balkans, Central Asia, and the Caucasus.
- I currently serve as Chief Technology Officer at Destek A.Ş., where I focus on secure, resilient, and scalable cloud and AI solutions, with particular emphasis on application-first architectures, cloud-native platforms, information security, high availability, and disaster recovery.
- I also teach applied artificial intelligence and large language models to Healthcare Management students.
- I support RePG Energy Systems as an early-stage investor.
My Blog
Bu web sitesinde çerez kullanılır.
Web sitesi trafiğini analiz etmek ve web sitesi deneyiminizi optimize etmek amacıyla çerezler kullanıyoruz. Çerez kullanımımızı kabul ettiğinizde, verileriniz tüm diğer kullanıcı verileriyle birlikte derlenir.