Matteo Craviotto

Production AI systems for investment workflows — multi-model orchestration, systematic trading platforms, and automation tooling between investors and their data.

View Projects Get in Touch
Los Angeles, CA · MS Financial Engineering @ USC · GPA 3.95
Scroll

About Me

Production AI systems for investment workflows — multi-model orchestration, systematic trading platforms, and automation tooling between investors and their data.

The pattern: take a manual, judgment-heavy financial process and replace it with something systematic and measurable. Running a multi-model trading system across 50+ US equities, a live prediction-market platform on Kalshi with a deployed risk engine, and AI orchestration infrastructure for investment research workflows.

Focused on where AI agents meet capital markets: investment research automation, systematic strategies, and tools that compress the distance between a thesis and a trade.

Download CV →

Professional Experience

Self-directed

Independent Quantitative Researcher & Developer

Sep 2025 – Present Los Angeles, CA
  • Designed and deployed a live risk engine for a prediction-market trading platform (Kalshi): exposure caps, daily loss limits, confidence-adjusted sizing, and Azure-hosted monitoring dashboards. Authored a 10-page forensic post-mortem decomposing live-vs-paper PnL divergence into component risk factors with pre-registered out-of-sample filters and shadow-data logging.
  • Built a multi-model AI orchestration framework in Python dispatching configurable analytical agents across leading LLM APIs — cost-efficient, self-improving infrastructure for automating research, monitoring, and reporting workflows. Proof-of-concept testing for investment risk automation: document summarization, data aggregation, anomaly flagging, and dashboard generation.

USC Marshall School of Business

AI Integration Consultant

May 2026 – Present · Los Angeles, CA
  • Helping Marshall faculty understand current AI capabilities and design controlled learning environments for students — hallucination-constrained setups where model outputs are scoped to professor-defined teaching policies (e.g. guiding students through reasoning rather than delivering final answers).

Graduate Teaching Assistant · FBE-529 Financial Analysis & Valuation

Jan 2025 – May 2026 · Los Angeles, CA
  • Taught equity valuation (DCF, multiples, LBO, APV, economic profit) to 30+ MBA students; graded weekly case memos and provided detailed written feedback on financial analysis and assumption quality across diverse industries and capital structures.

DefineX

Analyst · Data Quality & Process Automation

Apr – Sep 2024 Milan, Italy
  • Identified and quantified data quality gaps in loan-application document processing for major Italian financial services clients; designed a structured evaluation framework with 50+ tested variants, improving extraction accuracy from ~70% to 90%+ and eliminating manual review bottlenecks for credit risk teams.

Euronext Securities

Custody Intern · Income & Fiscal Services

May – Nov 2023 Milan, Italy
  • Reconciled 1,000+ annual corporate events (dividends, coupons, tender offers, mergers) across derivatives and asset-backed securities, identifying and resolving operational discrepancies across custody, tax, and operations teams. Built an automation tool cutting processing time from ~1 hour to under 1 minute — an operational risk control improvement with direct, measurable impact.

Deloitte Consulting

Junior Analyst · Financial Services Industry

May – Aug 2022 Milan, Italy
  • Designed a secondary-market tax-credit platform adopted by 5+ major financial institutions, translating investor workflow requirements into technical specifications enabling €10B+ in transactions.

Projects

Applied work at the intersection of AI agents, systematic trading, and quantitative research

Synapse — Multi-Model Autonomous Trading System

GitHub →

Python · Multi-LLM Orchestration · Alpaca · Docker · PostgreSQL

Autonomous trading system orchestrating five LLM APIs across 50+ US equities. Integrates real-time market data, volatility and regime signals, and model-consensus logic into a Dockerized pipeline with automated execution via Alpaca.

5Models Orchestrated
50+US Equities
24/7Dockerized Uptime

Kalshi Prediction-Market Risk Engine

GitHub →

Python · Kalshi REST API · Azure · PostgreSQL · Risk Analytics

Live risk engine for a prediction-market trading platform: exposure caps, daily loss limits, confidence-adjusted sizing, and Azure-hosted monitoring dashboards. Edge model empirically calibrated on 72M+ historical contracts. Forensic post-mortem decomposing live-vs-paper PnL divergence into component risk factors.

72M+Contracts
LiveAzure Deploy
LayeredRisk Controls

Multi-Model AI Orchestration Framework

GitHub →

Python · Multi-LLM Orchestration · FastAPI · Azure · Pydantic

Python framework dispatching configurable analytical agents across leading LLM APIs — cost-efficient, self-improving infrastructure for automating research, monitoring, and reporting workflows. Proof-of-concept testing for investment risk automation: document summarization, data aggregation, anomaly flagging, and dashboard generation.

AgenticWorkflow Engine
ParallelExecution
AzureProduction

Numerai ML Alpha Signals

GitHub → Details →

Python · LightGBM · Purged CV · Feature Engineering

Cross-sectional ML alpha signals on ~280M rows and 2.4k features for market-neutral 20-day return forecasts. Deployed live in Numerai's hedge fund with automated weekly retraining.

1.38Validation Sharpe
0.64Live Sharpe
300+Eras Tested

Macro Regime Duration Model

GitHub → Details →

Python · statsmodels · Markov-switching · Nelson-Siegel VAR

3-state Markov-switching model detecting latent Recession/Growth/Expansion regimes in GDP, inflation, and unemployment. Integrated yield-curve forecast with regime-conditioned mean-variance optimizer.

1.08Sharpe
−45%Max Drawdown
87%Persistence

Delta-Neutral Options Market-Making Simulator

GitHub → Details →

Python · Black-Scholes · LightGBM · cvxpy · Streamlit

Minute-level NBBO-aware delta-neutral simulator with BS theoretical pricing, Greeks calculation, and put-call parity residual tracking. Asymmetric spreads with inventory lean and delta-band hedging.

8–15%Target Fill Rate
550+Fills (800-min)

Options Parity Checker

GitHub → Details →

Python · Options Pricing · Real-time Data

Real-time arbitrage detection tool for put-call parity violations across options chains.

Post-Earnings Drift Backtest

GitHub → Details →

Python · Event Study · Statistical Analysis

Systematic analysis of post-earnings announcement drift patterns across market cap segments.

Get in Touch

Open to AI engineering and quant developer roles, collaborations, and discussions

Email

[email protected]

LinkedIn

/in/matteocraviotto

GitHub

@mcravi8

15-min Chat

Book a quick call →
Work authorization. F-1 visa, STEM OPT eligible — 3 years of work authorization from Dec 2026, no employer petition required. UK: EU Pre-Settlement Scheme, no employer sponsorship required.