Bitcoin LSTM Price Prediction
A PyTorch LSTM trained on order-book and market data with TA-Lib indicators, stored in SQLite and trained on Colab across multiple time resolutions.
Computer Engineering graduate from Ege University, currently a software development intern at Otokoç Otomotiv.
~/basak $ npm run dev · server ready in 312ms ▍
I graduated in Computer Engineering from Ege University in 2025 and I build web applications end to end — React and TypeScript on the front, Java Spring Boot and SQL on the back — with a real fondness for interfaces that feel good rather than just look fine.
I like carrying a feature all the way from database schema to the last pixel, and I'm happiest on teams where design and code talk to each other. Outside work: games and a permanently full coffee mug.
The tools I work with day to day.
Things I built, kept and still use.
A PyTorch LSTM trained on order-book and market data with TA-Lib indicators, stored in SQLite and trained on Colab across multiple time resolutions.
ER models of real booking platforms analysed and merged into one normalised MSSQL schema for multi-modal travel — relational design, normalisation, integrity rules.
Responsive storefront on Fake Store API: product listing and detail pages, Redux-managed cart persisted to localStorage, MUI drawer cart and global loading state.
Character discovery app with real-time filtering, multi-field sorting, pagination and detail views — dark theme, loading states and error handling throughout.
The site you're on: a bilingual (TR/EN) portfolio built with Next.js and TypeScript. Features an animated pixel-art scene drawn entirely in CSS, three switchable themes with no-flash persistence, a playful custom sword cursor, mouse-parallax motion and a Formspree contact form — plus full SEO with Open Graph and sitemap.
Spring Boot and Thymeleaf app with role-based access for doctors and patients, conflict-free scheduling, JPA-backed MySQL persistence and medical notes.