Projects
TicketLock • TicketLock.com
TicketLock is a sports ticketing platform my cofounders and I built to solve a problem we kept facing as fans. Tickets would go on sale months before tournaments, long before we knew who was playing. But if we waited, resale prices were outrageous or the seats were already gone. It felt like a lose-lose choice: commit early and gamble, or wait and overpay. At the same time, I was studying finance and engineering, spending every day with options and futures. That’s when the idea clicked: what if fans could reserve tickets the same way investors lock in outcomes? That’s what we built with TicketLock. TicketLock lets fans reserve face-value tickets for the specific matchups they actually want to see. By purchasing a non-refundable Lock, fans secure the right (but not the obligation) to buy a ticket if their chosen matchup happens. Locks can be structured for player vs field (like Alcaraz vs anyone) or for exact matchups (like Alcaraz vs Sinner), and we support similar logic across team sports such as basketball, soccer, american football, etc. Lock pricing is determined by a probabilistic model trained on supervised machine learning algorithms that ingest a wide range of inputs including tournament structure, player and team rankings, Elo-based performance metrics, surface-adjusted and condition-specific win probabilities, historical draw data, recent form, head-to-head statistics, and more. These features are used to estimate conditional matchup likelihoods, which feed into a dynamic pricing engine that mirrors financial option theory and adjusts premiums in real time based on implied volatility and path-dependent exposure. The result is a new layer of predictive pricing, structured exposure, and fan-level optionality in a ticketing market that’s traditionally reactive and speculative.
Mathify • Coming to market soon
Mathify (patent 63/723,212) is a physical keyboard that two close friends and I developed for people who work with mathematical notation every day. It is designed for students, researchers, and professionals in fields like engineering, physics, data science, and applied mathematics. We built Mathify while studying at Columbia, where we were constantly typing equations for assignments, lab reports, and research papers. Searching for symbols, recalling LaTeX commands, or formatting expressions manually in Word and Overleaf disrupted our workflow. We wanted a faster and more structured way to input technical content. Mathify features a custom physical layout with dedicated keys for symbolic notation, Greek characters, and complex mathematical operators. It enables structured input of advanced expressions such as nested integrals, multivariate summations, matrix algebra, and piecewise definitions through guided key sequences mapped to semantic logic. This allows for precise and consistent input across a range of technical environments. The hardware is fully programmable using QMK firmware and supports hot-swappable mechanical switches, letting users tailor both functionality and tactile feel. Mathify is paired with companion software that provides real-time formatting assistance and intelligent equation compilation powered by AI. The AI system parses input structure, auto-formats notation, and resolves ambiguous expressions based on context. The software integrates seamlessly with major technical platforms including Microsoft Word, PowerPoint, Google Docs, Overleaf, and LaTeX editors.