Software Engineering Analyst at Wells Fargo
Investigate and remediate security vulnerabilities across Java applications, tracing issues through application code and dependencies and validating fixes.
Gage Weaver
Software Engineering Analyst at Wells Fargo with experience in application security, full-stack development, machine learning, and algorithmic trading.
About Gage
I earned a degree in Computer Engineering from the University of Kansas and am pursuing an M.S. in Computer Science at Georgia Tech.
Most of my work is centered around applied machine learning and full stack development.
In my free time I enjoy learning about cars, collecting sneakers, and watching F1.
Professional experience
Software Engineering Intern at Wells Fargo
Full-stack services with Java, Spring Boot, and React. Improved observability for the Platform Engineering teams.
IT / Data Analytics Intern at Adams Brown
Power BI dashboards, DAX, SQL, and a tool that auto-generated email signatures so they weren't done by hand.
Education
B.S. Computer Engineering - University of Kansas
Machine Learning, Deep Reinforcement Learning, Data Science, AI, Computer Architecture.
M.S. Computer Science - Georgia Tech
Machine Learning for Trading and AI Computer Graphics.
Selected software and machine learning projects
IMC Prosperity Trading Challenge
A global top-16 finish in an algorithmic trading competition with more than 18,000 participants.
Designed and optimized algorithmic trading strategies across diverse assets and market conditions while solving open-ended market challenges under real-world constraints.
- 16th globally out of 18,000+ participants (top 0.1%)
- 6th in the United States
- Built strategies with Python, NumPy, and time-series analysis
Voyager
The Agentic AI Platform for Business Travelers. Book Your Travel, Navigate Delays, and More. Just Ask Voyager.
Voyager is a local AI agent that can book travel, navigate issues, and more. It uses a local Qwen model and tool-calling to interact with the user, the app, and the world.
- 2nd Place: Lockton Track at HackKU 2026
- An entirely local AI agent that can book travel, navigate issues, and more. It uses a local Qwen model and tool-calling to interact with the user, the app, and the world. Shipped in 36 hours.
PeTAI
Real-time form coaching for workouts, on iOS or in the browser.
PeTAI uses on-device pose estimation to count reps, flag bad form, and give plain-English coaching cues in real time, no server needed.
- 1st Place: Pella Corp. Track at HackKU 2025
- Web + mobile interfaces with real time vision and coaching shipped in 36 hours
CheckYoSelf
Authenticated P2P marketplace with 3D scanning, AR viewing, and on-chain holograms.
A peer-to-peer marketplace where every listing carries a 3D scan and a verifiable hologram, so buyers can inspect items as if they were holding them.
- 2nd Place: Midwest Block-a-Thon 2025
- Every listing contains an item history as well as the last state to verify authenticity.
- iOS + Web interfaces with real time 3D capture, AR preview, and IPFS interactions shipped in 24 hours
PocketDJ
Sentiment analysis turns any image into the perfect soundtrack.
Drop in a picture, get a song. PocketDJ runs vision-driven sentiment analysis on the image and queries the Spotify API for tracks that match the mood. Functions as both a web app and Adobe Express Extension.
- Winner: Most Creative Adobe Express Add-On at HackKU 2024
- Two delivery surfaces from one shared model
Kaggle Mental Health Classification
GPU-accelerated CatBoost finishing 7th of 2,600+ at over 94% accuracy.
A top-10 finish on a 2,600+ team Kaggle competition. Robust feature engineering + data preprocessing led to a strong finish.
- 7th of 2,600+ teams on final leaderboard
- >94% accuracy on the private set
- Single-model solution
Technical skills
Languages: Python, Java, JavaScript, SQL, C/C++
Machine Learning: PyTorch, Scikit-Learn, CUDA, TensorFlow, NumPy, Pandas
Web: React, Spring Boot, TypeScript, Node.js
Tools: Git, GitHub Actions, Jupyter, Docker