Can Polat
Can Polat

Machine Learning Engineer | Software Developer | Computational Scientist

Texas A&M University
JPu AI & Software Consultancy
can.polat0@yahoo.com

Howdy!

I am a Ph.D. candidate in Computer Engineering at Texas A&M University, focusing on the intersection of generative AI and materials science. My research employs cutting-edge techniques in computer vision (CV) and natural language processing (NLP) to advance material design and discovery. Additionally, I have served as a reviewer for prestigious conferences, including ICLR, KDD, and DSAA. I hold an M.Sc. in Physics from Boğaziçi University and a B.Sc. in Physics Engineering from Hacettepe University.

Beyond my academic career, I have been professionally active since 2019, successfully leading projects across diverse industries, including media, defense, and consumer electronics. In these roles, I deployed state-of-the-art machine learning models and computational imaging systems to solve complex, real-world problems effectively.

Curriculum Vitae
Interests
  • Machine Learning
  • Computational Physics
  • Materials Science
  • Computational Materials
Education
  • PhD, Computer Engineering

    Texas A&M University

  • MSc, Physics

    Bogazici University

  • BSc, Applied Physics

    Hacettepe University

Research Focus

I am a computational scientist working at the intersection of generative AI and materials science. My academic research explores advanced machine learning techniques, including supervised and unsupervised learning models for novel material discovery. I apply these methods using CV and NLP, particularly leveraging large language models (LLMs) to develop innovative solutions in materials science. I also utilize techniques like density functional theory (DFT) to support my work in materials design and property prediction. You can find my publications within the website or via my Google Scholar link.

While materials science remains a core focus, my interests are not limited to this field. I am also deeply fascinated by quantum lasers, where I have worked on integrating deep learning techniques for laser micromachining. Additionally, I am passionate about exploring the intersections of AI with various scientific domains, including drug discovery, energy storage, sustainability, medical imaging, diagnostic systems, and computational imaging, where machine learning can enhance accuracy and efficiency. My curiosity spans diverse areas of science, and I am always eager to expand my expertise into fields where AI can drive innovation and discovery.

Professional Experience

Over the past five years, I have gained extensive industry experience in machine learning and computational imaging. As a Senior Machine Learning Engineer at Wavebreak Media, I led projects focused on visual content generation and enhancement. I developed and deployed machine learning models for tasks like text-to-image generation, super-resolution, and 3D asset reconstruction. In this role, I used tools such as AWS, Google Cloud, Flask, Docker, MySQL, and OpenSearch to build scalable AI systems that optimized user experience and data retrieval for millions of assets.

Prior to this, I worked as a Computational Imaging Engineer at Aselsan and Arçelik, where I developed sophisticated optical systems for both defense and consumer electronics. I applied advanced computational techniques, including Fourier optics and ray tracing, to design imaging and display systems, further enhancing my expertise in computational methods and real-world engineering applications.

Featured Publications
Recent Publications
(2024). Multimodal Neural Network-Based Predictive Modeling of Nanoparticle Properties from Pure Compounds. Machine Learning: Science and Technology.
(2023). High-precision laser focus positioning of rough surfaces by deep learning. Optics and Lasers in Engineering.
(2023). A Transformer-Based Real-Time Focus Detection Technique for Wide-Field Interferometric Microscopy. 2023 31st Signal Processing and Communications Applications Conference (SIU).
Recent & Upcoming Talks