
About
I’m Jared Kramer, a Computer Science student at UW-Madison (’26) who loves turning ideas into fast, user-friendly web apps. I’ve built production UIs in React + TypeScript, shipped full-stack features with Express and Python, and worked across data and cloud-BigQuery, SQL, Google Cloud, and IBM Cloud. Recently, I was a React dev on an IBM team building enterprise UI components and pages, and I created an end-to-end analytics tool with a Python backend and React frontend for a retail brand. I also led an AI prediction-market project that combined Yahoo Finance data, Polymarket outcomes, and LLMs to drive 84% accuracy and 48% ROI. When I’m not coding, you’ll find me playing guitar, bass, or drums.
Projects
Gnosis Prediction Model
Complete
Built an AI trading agent for a Czech startup's prediction markets. The model combined Yahoo Finance data, Polymarket history, and web-scraped signals, orchestrated with LangChain + OpenAI, and calibrated with isotonic regression-ultimately becoming the company's #2 agent with 84% accuracy and 48% ROI.
Tech: Python (pandas, numpy, asyncio), LangChain, OpenAI, yfinance, Firecrawl (scraping), Polymarket/The Graph, scikit-learn (isotonic), JSONL logging.
Recz
In-Progress
A personalized music discovery app built with React, TypeScript, and Spotify API. Features swipe interface for discovering new music, ML-enhanced recommendations that learn from your preferences, and comprehensive listening history with advanced filtering by genre, decade, mood, and audio features. Includes playback controls, playlist management, and real-time feedback tracking.
Tech: React, TypeScript, Vite, Spotify Web API, Redis
Roller Rabbit Customer Analytics Map
In-Progress
An interactive geospatial analytics application that visualizes Shopify customer data on a dynamic map. Built with React and Flask, it integrates with Google BigQuery to analyze 12 months of e-commerce data, calculating customer lifetime value, purchase frequency, and category preferences. Features JWT authentication and machine learning clustering (KMeans/KPrototypes) to segment customers by spend, order patterns, and purchase diversity, with 24-hour caching for instant results.
Tech: React, Leaflet, Flask, Google BigQuery, scikit-learn, JWT
Triad Builder
Complete
A React-based music theory quiz game for learning and practicing triads. Features customizable difficulty with multi-select filtering, instant feedback, keyboard navigation, and a groovy retro interface. Fully responsive and optimized for mobile and desktop use.
Tech: React, CSS (animations, gradients, flexbox, media queries), JavaScript
Come Play!
Recognition
IBM Internship Program Certificate
IBM
Completed IBM's intensive software development program
View CredentialIBM Accelerate - Software Developer
IBM
Completed intensive training in software development fundamentals, GitHub, SDLC, algorithms, and agile methodologies with demonstrated coding proficiency.
View CredentialEnterprise Design Thinking Practitioner
IBM
Certified in IBM's Enterprise Design Thinking framework with expertise in user-centered design, empathy-driven ideation, and applying design thinking methodologies to real-world projects.
View CredentialIBM Agile Explorer
IBM
Foundational certification in Agile values, principles, and practices including Scrum, Kanban, sprints, and collaborative team methodologies.
View Credential
Contact
Or email me at jaredkramer321@gmail.com