Finance Lab | Established 2026

Investment systems that behave like a research desk.

Eleven systems spanning private equity, systematic strategies, and derivatives. Built, tested, and deployed as a unified decision framework.

Systems Live
10 / 11
Tests Passing
1,251
Current Phase
Capstone
Stack
Python, Next.js
01Systems
3 of 11 featured
Case 01 / CapstonePLANNED

Unified Research Terminal

Ten systems collapse into one research desk.

Live market data, financials, macro, and every engine in the lab. The AI Research Agent runs as the analytical backend.

Repo Reserved
Specification
Status
Planned
Integrates
10 prior systems
Stack
Next.js, Python, Plotly
Case 02 / EliteLIVE

AI Research Agent

Six engines decide. The LLM only writes the memo.

Deterministic recommendation aggregator. Engines run in parallel. The model writes the prose. The rating is locked.

View Repository
Specification
Tests
203
Engines
6, parallel
Composite
BUY ≥ 65, SELL ≤ 35
Stack
Python, Claude API
Case 03 / AdvancedLIVE

Portfolio Optimization Engine

HRP with RMT cleaning beats classical Mean Variance.

Four optimizers, three covariance methods. Ten year rolling backtest across fourteen ETFs. Sharpe 0.62 at 10 percent target volatility.

View Repository
Specification
Tests
166
Methods
4 x 3 grid
Best
HRP Sharpe 0.62
Window
10 year rolling
02Architecture
11 systems indexed

Eleven systems. Built to compose.

FoundationDeal modeling and screening.
Quant CoreFactor research and transactions database.
SystematicVol targeting, regimes, cross asset momentum.
AdvancedStrategy validation and portfolio construction.
EliteDerivatives and autonomous research synthesis.
CapstoneUnified terminal integrating the full stack.
#SystemTierTestsKey MetricStatus
01LBO EngineFoundation-Monte Carlo, 500 simsLIVE02PE Target ScreenerFoundation1490 companies scoredLIVE03Factor Backtest EngineQuant Core166Sharpe 1.55, 503 tickersLIVE04M&A DatabaseQuant Core124390 deals, 11 sectorsLIVE05Volatility Regime EngineSystematic109CAGR 10.9, Sharpe 1.03LIVE06TSMOM EngineSystematic10313 ETFs, cross assetLIVE07Strategy Robustness LabAdvanced136PBO via CSCVLIVE08Portfolio Optimization EngineAdvanced166HRP Sharpe 0.62LIVE09Options Pricing EngineElite2303 models, 8 GreeksLIVE10AI Research AgentElite2036 engine pipelineLIVE11Unified Research TerminalCapstone-Full integrationPLANNED
10 live / 1 planned1,251 tests passing
03About

Modeling and shipping the analytical layer of an investment firm.

François Rostaing | CFA Level I candidate

The Finance Lab demonstrates three signals at once: real financial understanding, strong data engineering, and decision tools investors actually use.

Each system ships with full test coverage and a Codex audited review. Each engine is reusable. The capstone integrates the prior ten into a single research interface.

Profile
RoleInvestment Systems Builder
CredentialCFA Level I Candidate
LanguagesPython, JavaScript, SQL
FrameworksNext.js, Streamlit, Plotly
DataDuckDB, SQLite, yfinance, SEC EDGAR
AIClaude API, deterministic orchestration
Lab Status10 / 11 live
Total Tests1,251
04Contact

Get in touch.

Open to roles in private equity, hedge funds, and quantitative research. Available to discuss any of the systems above.

Emailfrancois@frostaing.comGitHubgithub.com/FrancoisRost1LinkedInlinkedin.com/in/francois-rostaing