Flying Dutchman Labs

Flying Dutchman Labs was born from the frustration of seeing sophisticated narratives wrapped around fragile portfolios. The lab is an independent R&D initiative focused on portfolio construction first, where investment ideas are treated as inputs into a coherent optimization and risk framework rather than standalone bets.

We develop systematic strategies as well as selectively deploy fundamental, due-diligence-driven approaches. These research streams (systematic signal generation, newsflow analysis, and targeted fundamental investigation) feed into a unified portfolio construction and risk management engine, which remains the core of our work. Individual positions, whether long, short, or volatility-driven, are therefore expressed only in the context of their contribution to the overall portfolio.

Flying Dutchman Labs is not an asset manager but a research workshop. We do not manage external capital. We design strategies, tools, and portfolio construction frameworks that can be embedded into institutional investment processes via API, custom research, or white-label solutions.

Our work focuses portfolio construction, not stock picking. We design allocation engines using convex optimization and multi-objective methods (NSGA-III, MOEA/D and related algorithms) to search directly over the Pareto frontier of risk, return, drawdown, and factor exposure. Instead of treating risk as “just volatility,” we integrate tail metrics, scenario losses, and structural constraints (liquidity, borrow, turnover) into the optimization process. Investment ideas, whether systematic or fundamental, are evaluated only through their contribution to the overall portfolio. The result is a portfolio engineered to be consistent with the investor’s true risk budget, rather than reverse-engineered from a collection of ad hoc positions.

We develop systematic volatility strategies that integrate into the portfolio as both alpha and protection sleeves. In benign environments, they capture relative dislocations across single-stock options, index volatility, and term structures. As stress increases, the same framework tilts toward convex exposures and drawdown control, providing what we refer to as “chaos protection” - systematic exposure to asymmetric payoffs in dislocated regimes. Positioning and leverage are governed by the portfolio-level optimization and risk engine, so volatility exposures are calibrated relative to existing portfolio risks rather than evaluated in isolation.

Our equity L/S work is deliberately hybrid: argument-driven longs and forensic shorts. On the long side, we use systematic screening and LLM-based research agents to process newsflow, filings, and other primary documents, helping to form and update the qualitative thesis behind a position. On the short side, we focus on deep due diligence, including accounting quality, governance, capital structure, and business model fragility. Particular attention is paid to discrepancies between reported narratives and underlying fundamentals, which can form the basis for high-conviction, risk-aware short positions.

We believe every portfolio manager is first a risk manager, so the lab is built around coherent risk measurement and robust simulation frameworks. Portfolios are evaluated across a range of metrics under bootstrapped and Monte Carlo scenarios. We extend this approach with generative models to construct alternative market regimes and stress scenarios, allowing us to test whether a strategy’s edge persists beyond the sample in which it was developed. These outputs feed directly into the portfolio construction process, ensuring that risk is assessed dynamically and not inferred from past observations alone. This architecture is designed to continuously challenge our own assumptions before the market does.