HIGH ORBIT
ALT 1,200.0 KM
Applied AI Engineer Intern @ GE Aerospace · Full Time soon? 👀

Hi, I'm Jared 🚀

I build  

CS @ University of Cincinnati (Statistics minor). At GE Aerospace I build AI for aerospace (3× Corporate Impact Awards so far), and on the side I built DraftVision, an NFL prospect prediction platform that analyzes 9,000+ college prospects, with high-school prospects coming soon.

View my work → 📄 Résumé 🏈 Live: DraftVision LinkedIn GitHub
0
GE Corporate Impact Awards
0
Rework lead-time cut
0
prospects modeled (DraftVision)
0
users on tools I built
Selected work
Things I've built & shipped
Python React XGBoost CatBoost Docker

DraftVision — NFL Draft Prediction Platform

A full-stack ML app predicting NFL draft success for 9,000+ college & high-school prospects via a calibrated XGBoost/CatBoost ensemble (~73% held-out accuracy), live ESPN/CFBD APIs, and Claude Vision mock-draft import. Deployed on Docker/Fly.io.

Visit draft.jkrek.com ↗
DraftVision app screenshot
PyTorch Gymnasium RL

Deep RL Football Bot

A Deep Q-Network trained on real NFL play-call data to optimize drive-by-drive strategy in a custom Gymnasium environment, with animated drive visualizations.

View on GitHub ↗
LLM Agents RAG MCP

GE Metrology AI Toolkit

A production RAG assistant over engineering drawings & inspection instructions that cut rework lead time by 50% — built and owned end-to-end, deployed to GE's internal agent library.

Applied AI @ GE Aerospace
Where I've worked
Experience
GE Aerospace logo
GE Aerospace — Applied AI Engineer Intern
Dec 2025 – Present · Cincinnati, OH
  • Built & own a production RAG toolkit that cut rework lead time 50%; earned 3 Corporate Impact Awards.
  • Shipped agentic automation (MCP + HPC) that compressed ~6 months of work into 5 days.
  • Rolled out AI tooling to 30+ engineers; built QC-Calcite (200+ users, error-free in prod).
P&G logo
P&G Digital Accelerator — Data Science Intern
Aug 2025 – Dec 2025 · Cincinnati, OH
  • Built a computer-vision → surrogate-NN pipeline: +30% KPI-prediction speed, −15% downtime.
  • AI-agent workflows that saved ~10 hrs/week and lifted throughput 18%.
TQL logo
TQL — Data Science & Analytics Intern
Jan 2025 – Aug 2025 · Cincinnati, OH
  • Fine-tuned a production XGBoost pricing model, improving rate-recommendation accuracy 8%.
  • Built a PowerBI dashboard used by Sales VPs to guide coaching & strategy.
About
A bit about me

I'm a senior at the University of Cincinnati (B.S. Computer Science, minor in Statistics) graduating Fall 2026, and I build ML that actually ships. At GE Aerospace I earned three Corporate Impact Awards for applied AI work, at TQL I tuned an XGBoost pricing model in production, and I turned a football obsession into DraftVision, a deployed platform that scores 9,000+ college prospects. Off the clock, my football analytics YouTube channel has crossed 15,000 views.

Certifications
Certified & verified
click to launch