Quantitative Developer & Trader · AI Engineer & High-Performance Systems · Forward Deployed Engineer

I design, build, and run production systems at the intersection of quantitative finance, agentic AI, and real-time infrastructure. 6+ years in live markets. Former Argentine Navy officer. MIT MicroMasters (Finance & Data Science, in progress) · Oxford Algorithmic Trading Programme.

I care about systems that can be explained, validated, and reliably operated — not just code that compiles.

📍 Rosario, Argentina → Global (remote-first) 🌐 quant-studio.pages.dev · LinkedIn ✉️ catalaniclaudiodamian@gmail.com

• Strategy robustness diagnostics — overfitting, drawdown fragility, regime blindness
• Backtest-to-live gap analysis — slippage, funding rates, execution modeling
• ML-augmented signal engineering — XGBoost, LSTM, ensembles, walk-forward validation
• Real-time execution infrastructure — WebSocket/asyncio engines, order management, risk controls
• Agentic AI deployment — multi-agent workflows, MCP, Claude Code, business process automation
• Risk decomposition — Sharpe architecture, exposure monitoring, compliance logging

Real-time execution engine — crypto & prediction markets Built asyncio WebSocket execution layer integrating Binance, Kalshi, and Polymarket. Cross-venue normalization, order-type abstraction, heartbeat/reconnect state machine. Rejected REST polling due to latency variance; documented trade-offs in architecture notes.

ML signal pipeline — systematic crypto strategies End-to-end pipeline: ingestion → feature engineering → XGBoost/LSTM ensemble → walk-forward validation → live execution. Focus on out-of-sample discipline and regime detection, not backtest Sharpe theater.

Agentic automation — Bocho (Forward Deployed Engineer) Designed and deployed agentic AI workflows for logistics and operations clients. One deployment avoided 3 additional hires; another recovered 6 hours/week per manager. Built with Claude Code, MCP, and multi-agent orchestration.

Quant-Studio — independent research & trading AI-powered quantitative research and trading systems. Full ownership: data pipelines, model development, signal generation, portfolio construction, live execution, monitoring.

30+ verified certificates. All technical work and professional communication in English.

Quant & Trading: Python · NumPy/Pandas · Scikit-Learn · XGBoost · Keras/TF · PyTorch · VectorBT · Zipline · PyPortfolioOpt · ccxt · IBKR API · Binance/Kalshi/Polymarket APIs · PostgreSQL · Docker · Cloudflare

AI & Agents: Claude Code · MCP · LangChain · Multi-agent orchestration · Prompt engineering · RAG · Local LLM deployment

Engineering: asyncio · WebSockets · FSM execution engines · event-driven backtesting · CI/CD · Linux/Shell · Node.js · AWS (EC2, Lambda) · Vultr

• Upwork Top Rated Plus — 1,032+ hours · 100% Job Success · $60K+ earned
• $10M+ AUM served across consulting engagements
• 6+ years designing and running production trading systems end-to-end
• Max 3 concurrent consulting clients — depth over volume

• Building Quant-Studio — AI-native quantitative research and trading infrastructure
• Deepening agentic AI deployment patterns for finance and operations
• Finishing MIT MicroMasters in Finance and Data Science
• Open to Quant Developer / Quant Researcher / Quant Trader / FDE / AI Engineer roles and select consulting engagements

I document architecture decisions, trade-offs, and rejected alternatives in repo READMEs. If you want to see how I think — not just what I ship — start with the pinned repos and their design notes.