Financial Headline Sentiment Analysis
Neural Network built with PyTorch that classifies different market related headlines
CS & Economics — University of Minnesota
AI fluent software engineer dedicated to developing real products, automations, and more.
Open to internship opportunities
I'm a Computer Science and Economics student at the University of Minnesota (class of 2027) and an AI-fluent engineer — I build with agentic tools like Claude Code every day, and I care most about turning ideas into products that actually ship.
Recently that's meant a neural network that reads market sentiment from financial headlines, a trading model that paper-trades crypto on Polymarket, and this site itself — built end-to-end with AI tooling. I currently intern at The Walsh Group, running on-site IT for two data-center construction sites, and at UMN I'm active in Social Coding and founded the Consulting & Strategy Group.
Working with
May 2026 — Aug 2026
One half of a two-person on-site IT team serving two data-center construction sites with 5,000+ employees and a combined $13B budget. Brought 40+ new site trailers onto the corporate network with Cisco Meraki, onboard new hires with compliant devices, and act as frontline support for a largely non-technical workforce.
Jan 2024 — Present
Work in a team of 2–8 drivers coordinating arrival and departure logistics for busy events such as NHL games.
May 2021 — Jan 2023
Independently taught students in grades 2–10 topics up to Calculus I. Developed rapport with students' families, leading to being named in several positive reviews and increasing student retention.
Neural Network built with PyTorch that classifies different market related headlines
A trading model that applies Monte-Carlo Black-Scholes pricing to Up/Down contracts for BTC, ETH, SOL, and XRP, placing paper trades on Polymarket's 5-minute price markets. Built with OpenClaw and GPT 5.5. Not profitable yet — actively in development. Try the live paper-trading sandbox below.
An ML model for game-to-game performance analysis of NFL teams. Improved prediction accuracy by 13% by benchmarking Random Forest, KNN, SVC, and Gradient Boosting models against each other.
The fastest way to reach me is email — I read everything.
zsyedks@gmail.comOpen to internship opportunities · Minneapolis, MN