Quantitative Analyst @ Georgia Tech

Overview

I engineered an automated algorithmic trading pipeline to evaluate ETF performance, transitioning the committee’s investment strategy from manual fundamental analysis to a scalable, data-driven quantitative approach.

In the Spring of 2025, I joined the Georgia Tech Investment Committee, the largest student-managed fund on campus. I was selected for the Quantitative Analyst mentorship track, where the objective was to develop rigorous backtesting frameworks to validate trading signals before deploying capital. My specific focus was on building an automated "screener" that could ingest real-time market data and identify undervalued tickers for a $1M simulated portfolio, emphasizing risk-adjusted returns over pure speculation.

Learning and Growth

The most significant learning curve was translating statistical theory into executable code that could handle live market volatility. The first phase of my project involved mastering the Bloomberg Terminal API and cleaning the noisy financial data it provided. I realized quickly that a model that works in a controlled environment often fails in live markets due to "overfitting." This required me to pivot from simple regression analysis to building a robust Object-Oriented (OOP) framework in Python that could test strategies across various macroeconomic regimes (e.g., high inflation vs. growth periods).

A large portion of my time was spent engineering the backtesting engine. I wrote scripts to simulate sector rotation strategies, calculating metrics like Sharpe Ratio and Maximum Drawdown to assess risk. I collaborated with the committee heads to refine these models, learning how to defend my algorithmic assumptions to a non-technical audience. I eventually optimized the pipeline to reduce manual research time by 60%, allowing the committee to analyze hundreds of tickers in the time it previously took to analyze ten.

Takeaway

This role taught me the critical difference between data analysis and actionable intelligence.

Working with the Investment Committee pushed me to apply my Computer Science and Systems Engineering coursework to high-stakes financial problems. I learned that in quantitative finance, the code must be as resilient as the logic behind it. Overall, I gained end-to-end experience in the data lifecycle—from ingestion via API to signal generation—and developed the confidence to build systems that autonomously drive investment decisions.