Quantitative Finance Solutions
Claudio Catalani
AI-Augmented Quant Finance Consulting
View Services →
Who I Work With
Four types of traders.
One common problem.
$1M–$50M AUM. Lean teams building systematic infrastructure. You've got strategies in backtest that haven't been stress-tested against live conditions.
- Overfitting in strategy selection
- No out-of-sample validation framework
- Backtest Sharpe vs live Sharpe diverging
- Risk decomposition missing
Discretionary or systematic. Trading BTC, ETH, alts — dealing with regime shifts, volatility clustering, and exchange-specific execution gaps.
- Backtest ignores funding rates & slippage
- Strategy breaks during regime shift
- High drawdown relative to returns
- No ML-based signal enhancement
Speed-first teams running equity, futures, or crypto. You need fast turnarounds, actionable findings, and zero fluff in the output.
- Execution gaps eating strategy edge
- High-turnover models degrading live
- No systematic risk governance
- ML models overfitting to noise
Quant-grade oversight without a full-time hire. Risk modeling, portfolio stress-testing, and AI-integrated monitoring on a fractional basis.
- No in-house quant capability
- Allocation decisions not statistically grounded
- Risk models outdated or absent
- Vendor tools lack customization
Track Record
Results from live engagements.
All figures anonymized. Outcomes reflect production deployments across crypto, equities, and derivatives.Reduction in maximum drawdown after overfitting diagnosis and risk decomposition. Strategy retained positive expectancy while cutting tail risk.
Sharpe ratio improvement after statistical arbitrage restructuring. Out-of-sample validation confirmed signal persistence across regime shifts.
Full diagnostic turnaround from raw strategy data to structured PDF report with ML enhancement roadmap and prioritized optimization plan.
What I Fix
Six problems that kill systematic edge.
The most common structural failures I find across all client types — from solo crypto traders to institutional quant funds.Parameters tuned on in-sample data produce strategies that fail on first live contact. Your backtest is a history lesson, not a forecast.
All segmentsSlippage, funding rates, exchange latency, and market impact are systematically underestimated. Live performance diverges immediately.
Crypto-specificStrategies calibrated in trending markets collapse in mean-reverting regimes. No detection layer means no adaptation.
All segmentsPosition sizing that looks conservative on paper becomes catastrophic in correlated drawdowns. Risk is decomposed wrong — or not at all.
Funds & prop desksHigh raw returns masking volatile paths and negative Sortino ratios. Institutional-quality risk-adjusted metrics require architectural changes, not tweaks.
Quant fundsMost systematic traders use classical statistics when ML-driven feature engineering could materially improve signal quality and regime adaptability.
Crypto + QuantServices
Structured engagements, measurable outcomes.
Every engagement is scoped, deliverable-based, and focused on statistical robustness. Maximum 3 active clients at any time.A fast, high-impact quantitative assessment of your trading approach. I identify structural weaknesses, measure statistical robustness, and deliver a concrete optimization roadmap. Works across equities, crypto, futures, and multi-asset portfolios.
- Statistical evaluation (Sharpe, Sortino, MDD, volatility)
- Overfitting risk & curve-fitting detection
- Risk decomposition by factor and regime
- Crypto: slippage, funding rate & execution gap analysis
- ML feature enhancement suggestions
- 60-minute strategy review call
- Structured PDF report + action plan
- NDA-ready · ACH / SWIFT / USDC
Full implementation of dianostic findings. I redesign or rebuild the components of your system producing statistical drag — from feature engineering to risk architecture. Includes code delivery and performance validation framework. Crypto variant includes exchange-specific execution modeling.
- All Tier 1 deliverables included
- Feature engineering improvements
- ML model integration (XGBoost, LSTM, ensemble)
- WFO backtest validation framework
- Risk module redesign
- Code refactoring & delivery (Python)
- Performance monitoring setup
- NDA-ready · ACH / SWIFT / USDC
Fractional quant advisor for funds needing ongoing statistical oversight without a full-time hire. Monthly reviews, continuous AI research integration, and real-time regime monitoring structured around your existing operations.
- Monthly system performance review
- Performance diagnostics report
- AI / ML research integration
- Strategy refinement recommendations
- Market regime monitoring & alerts
- Priority async availability
- NDA-ready · ACH / SWIFT / USDC
Start with a call →
How It Works
From first call to deployed system.
30-minute call to understand your strategy, risk profile, and objectives. We determine fit and scope before any commitment.
You share strategy docs, backtest data, and live logs. I run statistical analysis using AI-augmented quant tools across your full system.
Structured PDF with findings, risk decomposition, and a concrete optimization roadmap. Delivered within 5–7 business days.
Tier 2 or 3: full build and optimization of the improvement plan. Code delivered, validated out-of-sample, and integrated into your workflow.
Positioning
This is not signal selling.
I work exclusively with traders and funds that take a systematic, analytical approach to markets.- Trading signals or alerts
- Speculative portfolio management
- Mass subscription products
- Retail-focused "magic systems"
- Guaranteed return promises
- Copy-trading infrastructure
- Black-box with no explainability
- Quantitative strategy diagnostics
- Statistical robustness validation
- ML-augmented risk modeling
- Systematic infrastructure engineering
- Out-of-sample WFO backtest frameworks
- Crypto execution gap analysis
- Fractional quant advisory for funds
About
Built on discipline, validated in markets.
I believe trading should be systematic, measurable, and statistically validated. Markets reward structure, not optimism. My work focuses on helping traders transform fragile strategies into robust quantitative systems that survive regime shifts and capital scaling. Before quantitative finance, I served as a Deck and Weapons Officer in the Argentine Navy — where precision, resilience, and methodical execution under pressure weren't optional. That operational foundation shapes every engagement.
Philosophy
Inputs must be quantifiable. Risk must be modeled. Parameters must survive stress tests.
Performance that doesn't survive forward testing is curve-fitting. Every system gets validated on unseen data.
Maximum 3 active clients. Every engagement gets full analytical attention, not templated output.
Technical Stack
Selected Credentials
Fintech & Quantitative Finance
Oxford: Algorithmic Trading Programme
MIT 15.455x: Mathematical Methods for Quantitative Finance
MIT 15.415.1x: Theory of Modern Finance I
MIT 15.415.2x: Theory of Modern Finance II
MIT 6.86x: Machine Learning with Python
BAI003x: Reinforcement Learning
ETFM2016x: Electronic Trading in Financial Markets
CS198.1x: Bitcoin and Cryptocurrencies
Computer Science
6.00.1x: Introduction to CS and Programming (Python)
6.00.2x: Computational Thinking and Data Science
ALGS200x: Algorithmic Design and Techniques
ALGS201x: Data Structures Fundamentals
ALGS202x: Graph Algorithms
ALGS203x: Algorithms for NP-complete Problems
NET04x: Advanced Algorithmics and Graph Theory
CSE 330: Operating Systems
Credentials alone don't build robust systems. What matters is application. My work translates statistical theory into practical trading diagnostics.
Contact
Start with an assessment call.
If you're serious about strengthening your trading system, book a 30-minute discovery call. We'll review your current strategy structure, risk profile, and objectives — no obligation, no sales pitch. Works for all segments: quant funds, crypto traders, prop desks, family offices.