I build machine-learning systems that turn large-scale, messy data into outcomes you can measure.
// quantitative trading · computer vision · LLM agents
Data scientist and ML engineer with 3+ years turning data into measurable business value. At an 8-figure e-commerce company I built marketplace insight engines and AI pricing systems across Amazon, Etsy and Shopify; earlier, as a research intern at TÜBİTAK — Turkey's largest research institute — I trained deep-learning models for medical imaging.
I'm now completing an M.S. in Business Analytics at Cal Poly Pomona on top of a B.S. in Electrical & Electronics Engineering. I like problems where a model has to earn its place by moving a number — accuracy, margin, ROAS, latency. Outside of work I teach AI on YouTube and Udemy and serve as an OpenAI ChatGPT Ambassador.
Shipped 20+ data and ML systems for an 8-figure retailer across 13 marketplaces — pricing, demand forecasting, A/B testing and LLM-driven trend detection on Amazon, Etsy and Shopify.
Built deep-learning models for medical diagnosis at Turkey's largest research institute, integrating an AI system into a clinical imaging device for professional use.
Programmed KUKA robots for synchronized operation on a truck mass-production line; hands-on PLC programming, pneumatics and basic hydraulics.
Filters EMA-crossover signals by classifying market behavior around support/resistance. Three models score each setup — 91%+ of filtered signals reach a 1:2 risk-reward.
Region- and line-based object counting on top of YOLOv7 with custom functions. My most-used open-source project.
Screenshots live TradingView charts via Chrome DevTools and reasons over them using your own Pine-Script logic in the prompt. Persistent memory, swappable vision models.
An agent crew where Claude orchestrates three DeepSeek workers passing tasks back and forth — built a Minecraft clone in ~5 minutes for about $0.05. Drop the folder into any project.
End-to-end license-plate detection and recognition built on YOLOv7.
Scrape an X/Twitter timeline and chat with the tweets through an LLM.
GPA 3.51. Graduate study in analytics, statistics and data systems.
GPA 3.1. Vice President, Undergraduate Research Club. Peer tutor in signal processing and neural networks. German C1.