Get to know more about my background and skills
I'm Yiğit Tilaver, an AI & Software Developer based in Istanbul. I studied Ship and Marine Technology Engineering at Istanbul Technical University and came to software through it — writing code for problems I first met as physical, mechanical ones. That order shaped how I work: I learned to respect real-world constraints before I learned to model them.
Most of my work sits where machine learning meets things that actually move. I have built computer vision pipelines for pose estimation of cameras and autonomous surface vehicles, 3D digital twin interfaces for operational data, and real-time protective equipment detection for industrial sites. Alongside that I automate the unglamorous parts — OCR-driven document processing, operational workflows, multi-language platform rollouts.
I work mostly in Python for models and data, and TypeScript with Next.js for the interfaces around them, with Docker and cloud deployment in between. What I care about most is that a system stays explainable: a model whose decisions can be traced is worth more than a slightly more accurate black box — which is why my trading system reports SHAP values rather than just predictions.
Analytical engineering foundation in marine systems, naval architecture and applied mathematics — the background behind my work on autonomous surface vehicles and industrial computer vision.
Second degree pursued in parallel, covering software engineering fundamentals, algorithms and database systems.
One semester of engineering coursework abroad at the Barcelona School of Nautical Studies (FNB).