Quant Library
An explainable quantitative research workbench for market data, risk metrics, scenarios, and model comparison. The full application uses a Python/FastAPI backend and Next.js frontend.
Read project briefQuantitative thinking. Creative building.
I’m Devin. I work at the intersection of finance, data, and AI—building systems that help people investigate, understand, and make better decisions.
Charlotte, NC · Building at Ballzatram
01 / Selected work
Independent projects from my workshop. Built with AI-assisted development, directed through product design, architecture, testing, and iteration.
An explainable quantitative research workbench for market data, risk metrics, scenarios, and model comparison. The full application uses a Python/FastAPI backend and Next.js frontend.
Read project briefA land-research workflow for screening parcels, comparing shortlists, tracing sources, and drafting diligence questions. Browser fallback available; advanced synthesis requires a backend.
Explore prototypeEconomics learned through decisions: a collection of simulations about markets, incentives, monetary policy, and strategic behavior.
Enter the arcade02 / Experience
From enterprise balance sheets to intelligent workflows, my work connects analytical rigor with practical engineering.
Read my resumeSr AVP — Quantitative Model Solutions Specialist
Building agentic investigation workflows and applied machine-learning systems, with experience in anomaly detection, rates research, and model reliability.
AVP — Quantitative Finance Analyst
Supported production forecasting for liquidity, capital, and stress testing, with a focus on root-cause analysis and automated validation.
AVP — Senior Asset Liability Analyst
Owned stress testing and balance-sheet forecasting for HTM and BOLI portfolios, turning scenario analysis into decision support.
03 / The project collection
A growing collection of tools, playable experiments, and ideas taking shape. Each project is labeled by its current stage.
24 projects
An explainable quantitative research workbench for market data, risk metrics, scenarios, and model comparison. The full application uses a Python/FastAPI backend and Next.js frontend.
Read project briefA land-research workflow for screening parcels, comparing shortlists, tracing sources, and drafting diligence questions. Browser fallback available; advanced synthesis requires a backend.
Explore prototypeEconomics learned through decisions: a collection of simulations about markets, incentives, monetary policy, and strategic behavior.
Enter the arcadeAnalyze uploaded historical prices for return, volatility, drawdown, concentration, correlations, and risk contribution, with calculations performed locally.
Open portfolio labTranslate macroeconomic shocks into portfolio factor sensitivities with explicit assumptions and stress scenarios.
Run a scenarioAn earlier macro dashboard concept, now retained as a compatibility route into Quant Library.
View legacy routeA weather-market worksheet for settlement review, uncertainty, and capped sizing. The public worksheet does not place live orders.
Open worksheetA research workflow for comparing market behavior around selected events, with event windows and analytical assumptions made explicit.
View implementationCompare analytical model choices, validation signals, and stability tradeoffs before drawing conclusions.
View implementationPrepare selected project context for your own AI app. The Supply & Demand connector adds an interactive lab in compatible apps once activated; the connection guide shows its current availability.
Explore AI guideTurn saved analysis into editable Markdown reports that retain assumptions, provenance, caveats, and next checks.
Build a reportA short-form video editing workflow for owned or licensed media. Actual rendering requires the Python backend and media workers.
Explore editing workflowExplore equilibrium and welfare through taxes, price controls, shocks, predictions, and sandbox experiments.
Open market labTake the chair of a fictional central bank and navigate inflation, policy lags, credibility, and financial stability.
Take the chairExperiment with cooperation and betrayal against repeated-game opponent archetypes, then inspect payoffs and strategy debriefs.
Test a strategyA game-theory learning studio spanning auctions, signaling, bargaining, rational choice, and mechanism design.
Open strategy studioA market-clearing simulation exploring decentralized decisions, price signals, and policy feedback.
Run a marketAn absurd interactive-fiction experiment told through a fictional police blotter and branching decisions.
Play the storyA creative game world combining a playable manuscript, relic archive, hymns, and a rhythm-battle experiment.
Explore project sourceGenerated-story experiments exploring structured narrative workflows and editorial review.
Read architecture notesFollow bill activity, inspect official legislative records, and examine selected provisions with their sources. The broader promise and media-accountability work remains in development.
Explore bill trackerA persistent AI companion and learning vision: guide critical thinking through conversation, games, and experiences that grow with the learner.
Exploration in progressA reading companion exploring book tracking, recommendations, and a reader identity that develops as books are completed.
Exploration in progressAn Osiris concept connecting genealogical research and source-backed family graphs to historical narratives in a reader’s chosen style.
Exploration in progressNo projects match this search.
04 / A little context
I’m drawn to questions that sit between disciplines: how markets behave, how people learn, and how intelligent systems can turn information into something useful.
Ballzatram is where I explore those questions in public. It’s a workshop for quantitative tools, economic games, creative experiments, and the next idea worth building.
Education
University of Maryland · 2022
Texas Christian University · 2019
Beyond the workbench
Treasurer — Illuminating Engineering Society, Charlotte
Polo Manager — Camden Polo Club
Let’s talk about quantitative systems, AI workflows, or an idea that needs to become a working product.