Matthew Casertano
I'm a student at Caltech studying Computer Science and Economics, driven by curiosity about how data reveals hidden patterns. When I find something interesting, I build tools to explore it deeper.
Things I've Built
I'm drawn to projects that reveal hidden patterns or solve real problems. Here's what I've been working on:
🌡️ Climate Comfort Index
Ever wondered where in the US matches your ideal weather? This web app takes your climate preferences and maps them across 3,000+ counties, revealing patterns about where you'd actually want to live.
How it works: Input your ideal temperature ranges, humidity preferences, sunshine needs, and snow tolerance. The algorithm calculates a "comfort score" for each county and generates a beautiful choropleth map.
Fun fact: My model found that Pacific Northwest counties like Hood River, OR score highest for people who like seasonal variation without extreme temperatures.
🎮 Try the Interactive Demo
Want to see where you should live? Try it yourself (tip - reload the page if it doesn't work):
🗺️ Find Your Climate Match📈 Risk Parity Portfolio Analysis
Most portfolios are dominated by stock risk. What if we allocated based on risk contribution instead of dollars? I tested this contrarian approach across 18 years of market data, including three major crises.
Key Finding: A Stock-Bond-Gold risk parity portfolio achieved a 0.98 Sharpe ratio while limiting maximum drawdown to just 20.2% - dramatically outperforming traditional allocations during market stress.
Inspired by Bridgewater: Building on Ray Dalio's risk parity framework, I wanted to test how this approach holds up recently, especially given critics arguing that risk parity hasn't performed well lately. However, it proved particularly effective during both the 2020 COVID crash and 2025 tariff crisis, especially when gold was added as a third asset.
🇨🇳 China's AI Strategy
While everyone focused on the chip race, we identified how China could play a different game: leveraging open-source AI and legacy chip dominance instead of matching US semiconductor capabilities. Our timing proved prescient.
Core Insight: Rather than chasing TSMC's 3nm chips, China is positioning to control 40% of legacy chip production by 2027.
The Strategy: Three pathways to AI dominance: pro-AI legislation, open-source innovation, and legacy chip market control.
✅ Thesis Validated: One month after our research, DeepSeek shocked the market in a way that matched our predictions: breakthrough AI performance with legacy H800 chips and open-source strategies.
🏆 ML Competition Winner
First place out of 58 teams in Caltech's CS 155 Machine Learning Competition. The challenge: predict telecom customer churn with noisy, real-world data.
The Secret Sauce: While others focused on fancy algorithms, I went deep on feature engineering. Created interaction terms, polynomial features, and domain-specific transformations that captured hidden patterns in customer behavior.
Result: Beat graduate student teams and advanced ML approaches with thoughtful data preprocessing and feature selection.
🧬 NMRSuite
Built during my time at UMD's Fushman Lab - a web gateway for analyzing protein dynamics through NMR data. Took complex molecular modeling algorithms and made them accessible to researchers worldwide.
Impact: Published in Gateways 2021 and presented at the NSF-funded SGCI Conference. Now used by structural biology researchers to understand how proteins move and function.
Technical Challenge: Translating cutting-edge research algorithms into a user-friendly web interface that handles complex scientific data processing.
⚡ Personal Automation Tools
Small frictions in daily life add up. I built these simple Apple Shortcuts to eliminate the annoying stuff and make my day run smoother while I wait for Apple Intelligence to catch up.
Speed Bump: Forces me to articulate why I'm opening an app before it launches. Turns mindless browsing into intentional decisions and gets great data.
Add Event: Add calendar event from plain text (like forwarded emails) using ChatGPT to parse the details automatically.
Get Availability: Scrapes my calendar and returns available time slots over the next X days. No more manually checking availability.
🧠 Neural Autopilot Parameter Analysis
Parameter space analysis for a novel behavioral model explaining power laws in human behavior. Validated the potential for ground truth recovery of these models as part of research in the Colin Camerer lab at Caltech.
Research Impact: Contributed to understanding how habit formation and goal-directed behavior create bursty patterns in human activities like gym attendance and social media usage.
Technical Achievement: For robust parameter recovery algorithms, conducted stability analysis across 1000+ parameter combinations, and developed visualizations to understand these results.
Presentation: Presented findings to research team, focusing on simulation limitations and parameter space exploration methods.
Education
GPA: 4.1/4.0 • Research under Professor Colin Camerer on power laws in human behavior and LLM behavior in experimental markets • Selected for Housing Working Group to redesign campus housing lottery • Bechtel Steward representing student needs • Active in Quantitative Finance Club, Poker Club, Sunday League Soccer
GPA: 4.82 • SAT: 1600 • Captain of Math Team & Financial Literacy Club • Lecturer for Computer Team & Debate Team
Experience
Joining the world's largest hedge fund to work on systematic investment strategies and economic research.
Conducted due diligence on 20+ AI/ML and enterprise SaaS investments. Built automated sourcing systems and led projects on edge computing and AI for customer service, presenting findings to executive leadership.
Developed NMRSuite for protein dynamics analysis. Published research and presented at NSF-funded conferences. Bridged the gap between cutting-edge science and practical tools.
Built nonprofit from scratch during COVID-19, connecting teen volunteers with vulnerable seniors needing grocery deliveries and companionship. Scaled to 31 chapters across the US and Canada with hundreds of volunteers making nearly 4,000 deliveries over three years.
Media Coverage: Featured in major outlets including CNN, Good Morning America, NPR, The Washington Post, People Magazine, and PBS. Demonstrated how young people could organize effective community responses during crisis.