Jul 2025 – Jan 2026 · 6 mos · New York, New York, United States

• AI & ML Team (AIA Labs) — Bridgewater’s in‑house AI research arm building proprietary machine‑learning, reasoning‑engine, and LLM systems to drive alpha in macro investing

Jul 2024 – Jul 2025 · 1 yr · New York, New York, United States

• AIA Labs

Feb 2024 – May 2024 · 3 mos

Sep 2023 – May 2024 · 8 mos · Cambridge, Massachusetts, United States · On-site

• Developed an innovative Scientific Machine Learning (SciML) methodology to reduce drift in Inertial Navigation Systems (INS), enhancing navigation accuracy and reliability in GPS-denied environments
• Integrated neural network models with traditional INS components to effectively capture complex error patterns and adapt to dynamic conditions
• Conducted extensive simulations using Julia and the High-Performance Inertial Navigation Development Repository (HIDR) to generate realistic datasets and evaluate the performance of the SciML approach
• Demonstrated a 62.99% reduction in 3D RMSE and significant improvements in attitude estimation compared to traditional Kalman filtering techniques
• Applied advanced concepts from rocket science and navigation to push the boundaries of inertial navigation technology
• Contributed to the field of autonomous navigation, with potential applications in aviation, robotics, and aerospace systems
• Technologies: Julia, Neural Networks, Extended Kalman Filter, Inertial Navigation Systems, Scientific Machine Learning

Jun 2023 – Aug 2023 · 2 mos · New York, United States · Hybrid

May 2022 – May 2023 · 1 yr · Cambridge, Massachusetts, United States

• In this groundbreaking research project, we delved into the potential of Large Language Models (LLMs), specifically GPT-3, to enhance the decision-making process in cybersecurity scenarios. Our goal was to ascertain the viability of using LLMs for tasks such as anomaly detection, vulnerability identification, and attack forecasting within a cybersecurity context.
• The study introduced an innovative experimental setup, employing a graph-based cybersecurity network, where GPT-3 played a pivotal role in influencing the decisions of a defensive agent. The methodology we used involved complex simulation environments that mimicked real-world cyber threat landscapes to rigorously assess GPT-3's capabilities.
• The initial findings were promising, highlighting the LLM's proficiency in devising optimal paths and strategies in the simulated settings. The project not only yielded a more comprehensive understanding of LLMs' role in cybersecurity but also provided valuable insights with the potential to substantially advance cybersecurity methodologies and tools.
• As the leading researcher, I was responsible for conceptualizing the study design, executing the simulation-based experiments, analyzing the resulting data, and interpreting the outcomes, all of which led to a published research paper. This research represented a significant leap in utilizing AI's potential to bolster cybersecurity defenses and optimize decision-making processes.

May 2022 – Aug 2022 · 3 mos

• Created a risk framework for a new protocol being launched by Delphi Digital
• Analyzed various statistical and machine learning techniques to help create a more quantitative approach to assessing risk
• Researched ways to better assess different decentralized finance risk parameters

Feb 2022 – Sep 2023 · 1 yr 7 mos

Feb 2021 – Feb 2022 · 1 yr

• Direct a coding competition where participants learned machine learning concepts to create a poker bot

Sep 2020 – Feb 2021 · 5 mos

• • Campaigned for over $80,000 in company sponsorships and managed club expenses

Feb 2022 – Aug 2022 · 6 mos

• Analyzing recent developments in how Transparency and Portfolio Trading is transforming the Corporate bond market
• Performing traditional Quantitative methods and strategies on newly released data sources

May 2021 – Aug 2021 · 3 mos · Chicago, Illinois, United States

• Developed Python application to parse thousands of financial lists from textual data sources
• Automatically mapped application to Bloomberg’s API to identify new mortgage-backed securities
• Aggregated and cleaned a dataset of over 12,000 mortgages to identify new trading opportunities

Mar 2020 – Feb 2021 · 11 mos · Greenwich, Connecticut, United States

• Researched a trading strategy by analyzing the correlation between 10,000 intraday stock prices and closing cross
• Engineered a script to monitor stock splits, earning reports, and daily fees for different U.S. equities
• Built an automated report by querying fee data across the major U.S. Market Exchanges such as NYSE and NASDAQ

Oct 2019 – Jan 2020 · 3 mos · Cambridge, United States

• Analyzed consumer preference data to create quantitative metrics that were used to understand consumer psychology
• Compared consumer sentiment between traditional and automated wealth management solutions