Python 
JavaScript 
React 
Node.js 
TypeScript 
Pandas 
Polars 
Spark 
C# 
Snowflake 
FastAPI 
MySQL 
PostgreSQL 
Docker 
AWS 
Git 
Linux 
MongoDB Python JavaScript React Node.js TypeScript Pandas Polars Spark C# Snowflake FastAPI MySQL PostgreSQL Docker AWS Git Linux MongoDB About Me
Currently based in New York City. I hold an MS in Operations Research from Columbia and a BS in Math & CS from École Polytechnique. Over the past 8 months, I've been working as a quantitative developer at AlphaQuest, a fully systematic hedge fund located on 59th and Park. I built systems for research and trading, see projects section to learn about some of the work I did there. Outside of work, I enjoy training Martial Arts (Jiu Jitsu / Boxing) and reading.
Awards
Winner - Amazon Bedrock Innovation Challenge
2025Won 'Most Scalable Solution' and $5,000 in AWS credits at the Amazon Bedrock Innovation Challenge with teammates Alessandro Massaad and Martina Paez Berru. Built an AI platform powered by Bedrock, using Agentic AI and RAG on 1,000+ publications, to let anyone ideate, codify, and backtest quant systematic trading strategies.
Excellence Scholarship
2021, 2022, 2023Awarded Excellence Scholarship by the École Polytechnique Foundation for three consecutive years. Excellence Scholarships are awarded to the very best students who demonstrate outstanding academic abilities, strong motivation, and a pioneering spirit to be real ambassadors of the Program.
Mathematics and Chemistry Olympiads
2019, 2020Competed at national levels in Mathematics and Chemistry Olympiads.
Education
MS in Operations Research
Columbia University
- Relevant Coursework: Computational Portfolio Construction, Model & Trade Derivatives, Fixed Income, Machine Learning, Deep Learning, Optimization, Simulations, Statistics, Probability, Algorithmic Trading, Industry Projects
BS in Mathematics and Computer Science
Ecole Polytechnique
- Relevant Coursework: Algebra, Analysis, Measure Theory, Topology Probability, Stochastic Processes, Statistics, Distributed Computing, Functional Programming, Object-Oriented Programming, Computer Architecture, Algorithms
Experience
Quantitative Developer Intern
AlphaQuest (Systematic Quantitative Hedge Fund)
- Quantitative Systematic Hedge Fund
Teaching Assistant
Columbia University
- IEORE4576: Data-Driven Methods in Finance, by Prof. Charles Pehlivanian
Machine Learning Engineer
ask2.ai
- GNNs for Portfolio Optimization, supervised by Prof. Ali Hirsa
Research Assistant
École Polytechnique
- Bachelor Thesis in Optimization: Grade A
Summer Associate
DNB Asset Management
- Renewable Energy Fund (Equities)
Projects
Pulse - Surveillance Platform
AlphaQuest (Systematic Quantitative Hedge Fund)
As a step towards alleviating the overnight trader role, built a full-stack surveillance platform used by trading, research, and engineering teams. Designed Snowflake data models and engineered a modular C# scheduling/alerting system with clean separation of executors, making new alert types plug-in without touching core logic. Developed a fast React UI and added manual execution, Slack/Datadog/PagerDuty routing, and health-state analytics—resulting in a more reliable, scalable, and easy-to-extend monitoring stack.
Athena - PnL Dashboard
AlphaQuest (Systematic Quantitative Hedge Fund)
Integrated Equities into internal PnL platform. Added processing for equity + Barra factors. Delivered new React UI tab with factor heatmaps and time-series views. Implemented feature- and contract-level PnL aggregation. Stack: Python, Snowflake, React, AG-Charts.
Trader Commentary Automation
AlphaQuest (Systematic Quantitative Hedge Fund)
Automated daily trader commentary, cutting ~20% of manual workload. Aggregates internal trading data, risk metrics, flow analytics; ranks key PnL/vol drivers; enriches with live market news. LLM with domain prompts drafts concise commentary aligned with house views. FastAPI microservice, REST architecture, fully integrated into trading stack. Built tooling so traders self-maintain prompts and data sources, improving quality week over week. Stack: Python, FastAPI, internal data pipelines, LLM prompt engineering, news scraping.
QuestGPT - Talk to your data
AlphaQuest (Systematic Quantitative Hedge Fund)
Production chatbot for PnL analytics. Natural-language queries → SQL generation, data retrieval, charts, AG-Grid tables, and text summaries grounded in internal commentary and market context. Multi-agent workflow: • SQL-Gen Agent: safe SQL for PnL, exposures, attribution • RAG Agent: internal commentary retrieval • News Agent: market context relevance • Orchestrator: routing + output fusion Fully integrated into trading stack. Eliminated most ad-hoc PnL requests. Stack: Python, FastAPI, AG-Grid/Charts, Snowflake, multi-agent LLM.
Alternative Data Parsing Pipeline
AlphaQuest (Systematic Quantitative Hedge Fund)
Designed and deployed a production pipeline extracting structured insights from unstructured Futures & FX flow delivered via email. Ingests attachments via secure email APIs, enhances charts (contrast, normalization, upscaling), detects numeric regions, and parses tables using LLMs with domain templates. Implemented validation with structural invariants (row/column consistency, numeric/unit checks) and auto-retry correction rules — ~99% extraction accuracy. Integrated with Snowflake for downstream analytics and automated trader commentary. Added a Snowflake view aligning weekly street-flow data with internal flows for apples-to-apples signal research. Stack: Python, image preprocessing, LLM parsing, email APIs, Snowflake.
GNN for Portfolio Optimization
Columbia Engineering
Built GNNs (GAT and GCN) learning portfolio weights from asset graphs. Backtested vs. MVO and Equal Weight. Key findings: Complexity > reward. Results unstable; interpretation weak. Risk gains offset by return drag. Constraints + noise → trivial allocations. Attention helps structure, not edge. Conclusion: strong theory, weak practical payoff. Needs better graph construction, interpretability, and regime adaptation. Stack: PyTorch Geometric, CVXPY, Python.
LLMs for Investment Ideation
Columbia Quantitative Research Labs, supervised by Prof. Micheal Robbins
Laid foundations for Columbia Engineering's Quantitative Research Lab. Architected modular pipeline processing 3.41 GB of financial research, unifying 5,000+ papers and 20,000+ posts with semantic search. Reduced research time by >100 hours/month for portfolio managers.