Joseph F. Ghanimah
I'm an engineer specializing in the intersection of machine learning, data science, and statistics. I build intelligent systems that solve complex problems, from developing predictive trading algorithms to architecting full-stack applications that serve ML models. My passion is translating theoretical concepts into practical, high-performance software, built on a foundation in Computer Engineering from the University of Illinois at Urbana-Champaign (UIUC).
Machine Learning & AI
Deep neural network predicting MOBA match outcomes purely from 10-champion team compositions using custom embedding layers trained on high-ELO games.
A custom-trained Large Language Model integrated into a Discord community with multi-layer query routing, context management, and personality tuning.
AlphaZero-inspired reinforcement learning agent combining dual-headed policy and value networks with Monte Carlo Tree Search lookahead.
Quantitative & Rating Systems
An enhanced team-based Glicko-2 rating system adapted for live per-game updates, roster strength weighting, and non-linear score margins calculated via Binomial CDF.
A multi-strategy algorithmic trading bot executing quantitative strategies live: leveraged ETF hedging (TQQQ/VIX), pairs trading cointegration, and automated market making.
Web Platforms
This personal website showcases my engineering work and skills in Python (Flask), HTML, CSS, and JS, featuring a custom 3D parallax depth layer.
A platform for collegiate Call of Duty esports, exploring tournament management, automated matchmaking, and community engagement.
