I blend advanced mathematics, machine learning, and full-stack engineering to turn complex ideas into real-world tools.
I'm a second-year student in M.Sc physics and mathematics, specializing in industrial mathematics. I build full-stack tools that sit directly on top of deep math, whether it's a real-time arrythmia classifier, a quant backtester, or a LaTeX-aware AI pipeline.
In 2024, I was diagnosed with heart failure. While recovering, I kept studying, training, and coding, often while dealing with irregular heartbeats that showed up on my own ECG data. That experience pushed me to focus on building systems that actually matter. Tools that monitor, predict, or automate something critical in real life.
I work mostly in Next.js, Python, and C++, and my current focus is split between real-time arrhythmia detection, cardinality-constrained portfolio optimization, and high-frequency crypto strategies using custom indicators and BTC context logic. I'm here to solve hard problems, not polish pitch decks.
A real-time ECG monitoring application that streams live cardiac data from Polar H10 chest straps and performs intelligent arrhythmia detection using advanced signal processing algorithms. Tracks PVC burden, detects patterns like bigeminy and trigeminy, and supports session-based analysis with planned alarm functionality for high-risk events.
This project explores portfolio optimization under cardinality constraints, where the number of assets selected in a portfolio is explicitly limited. Using integer programming, the project demonstrates how adding a cardinality constraint transforms a convex Markowitz optimization problem into a combinatorial, non-convex challenge.
A custom algorithmic trading system for Solana using custom signals, enhanced with BTCUSDT price context for confirmation. The strategy includes dynamic stop-loss, ATR-based take-profit, slippage modeling, and realistic funding/fee handling. Backtested on 5-minute candles with live simulation support. It has been deployed live and generated consistent profit in high freq trading environments.
A custom-built coaching website for a personal trainer, featuring a clean and responsive UI with registration and contact functionality. The platform serves as a digital entry point for potential clients to sign up for personalized coaching plans.
A startup discovery platform that delivers AI-generated business ideas daily. Users can explore, analyze, and expand on startup concepts. I contributed to the frontend architecture, user interaction flow, and content generation logic.
A web platform and Chrome extension designed to enhance and manage AI prompts more effectively. Prompt Fox allows users to save, organize, and inject optimized prompts directly into AI tools. I contributed to both the frontend and backend, as well as the core Chrome extension logic.
The society that separates its scholars from its warriors will have its thinking done by cowards and its fighting by fools.— Thucydides
Working as a fullstack developer in a venture studio environment, building and scaling early-stage startups from idea to launch. Responsible for frontend and backend development across multiple products, including AI-powered web apps, browser extensions, and MVP tooling. Collaborate closely with founders, designers, and product teams to ship rapidly and iterate fast.
Tutored high school students in mathematics, focusing on problem-solving and conceptual understanding. Developed personalized learning plans and resources.
Served as a combat engineer. Developed skills in teamwork, leadership, and problem-solving under pressure.