ironcrypto / abides Goto Github PK

• 🏢 20y-Exp Quant/Quant Dev | Trading, Research & Execution in TradFi-DeFi | Montreal, Canada
• 💫 PhD | Mathematical Finance | Paris Sorbonne, France
• 🌟 MSc Economics | Paris Sorbonne, France

Seasoned financial engineer and product innovator with extensive experience in capital markets and banking. I specialize in agent-based and machine learning applications, with a recent focus on crypto markets and decentralized finance (DeFi) where I have been operating for the last 2.5 years.

My expertise includes designing and implementing ModelOps platforms, quantitative modeling, trading execution system and leading projects in Model risk management, pricing, and regulatory compliance. As an entrepreneur and R&D leader, I excel in building new business opportunities, managing relationships, and driving innovation. I have a broad network in DeFi, FinTech, and Academia.

Known for my collaborative approach and ability to thrive in challenging environments, I'm a self-starter, early adopter, and effective communicator. I enjoy exploring new ideas, engaging in debates, and continually learning from others.

• Focus on LOB dynamics, market impact models, and TCA.

• CPO-ed, an AI agent-based platform destined to Quant Dev users to import, build, test, deploy and monitor autonomous agents running on the blockchain (EVM mostly) to execute low to mid-frequency DeFi strategies relying on AI-ML models in a non-custodial way. Led the Quant and Product teams for 2y, 0-to-1, product-market fit and go-to-market
• Collaborated with Igor Halperin's team and Professor Petter Kolm.
• Designed an Optimal Order Execution system using RL and MLOps design patterns, aimed at building a trading-as-a-service platform for institutional clients (details available on request).
• PO-ed and GTM-ed a ModelOps platform, cloud and open-source based for MLC operationalization for TradFi Institutional clients.

• Founded an algo trading tech and research firm.
• Built an aggregated order book for Europe's main MTFs, simulating up to 1 million orders/second.
• Analyzed MiFID best execution principles in high-frequency markets using C++ and Java for over 50 different order types. (repo provides a glimpse).

• Developed systems in Java to study HFT systemic impacts for regulation purposes.
• You can check this executive deck presenting the PoC-stage project.

• Modeling the quantum-chaotic nature of microstructure to refine limit order patterns and maximize MM rebates (work in progress).
• Building a na agent-based and RL-based backtesting engine for market-making strategies on digital assets.

Python	Rust	C++	MATLAB	Git	Jupyter	Linux
Bash	SQL	TheGraph			PyTorch	Kubernetes