AI has driven down the cost of knowledge and routine-problem solving. The skill that remains scarce is decision making. At SVVVL, we do research and build tools to enhance the quality of your decisions.
An evolutionary foundation for continuous-time asset allocation.
We extend the Brennan–Lo evolutionary framework to continuous-time asset allocation by shifting the unit of selection from investors to individual dollars. The growth-optimal strategy is randomization — a Kelly–Merton deterministic core plus a stochastic term governed by belief alignment.
SVVVL (pronounced "swivel") is a Decision Intelligence (DI) research, consulting, and engineering firm. We help clients turn their data into insights and those insights into goal-based improvements.
AI and ML are powerful forces that can accelerate decision-making. Decision Intelligence enables us to discover the most appropriate direction in which to apply those forces. Every engagement begins with understanding the context — goals, constraints, alternatives, and potential impact — before recommending any solution.
We start by understanding what success looks like for you, before touching the data.
Resources, timelines, risk tolerance, and regulatory factors all shape the solution space.
We map the landscape of possible approaches before converging on a recommendation.
We model potential outcomes to ensure solutions are practical, measurable, and defensible.
We draw on tools from multiple disciplines, because the best solution to your problem might live in any of them. Our research interests range from AI/ML to economic modeling and quantitative finance.
Structured analysis of your decision landscape before any AI/ML solution is proposed. DI delivers a quality decision.
SVVVL's FDE delivers AI-enhanced software that implements your DI .
From model selection to deployment — identifying where AI creates genuine leverage, and where simpler methods work better.
Rigorous quantitative models grounded in economic theory and statistical practice, tailored to your system's dynamics.
Applied research at the intersection of financial theory and modern ML — risk modeling, systematic asset allocation, portfolio construction, and alpha generation.
Deep-dive research for clients tackling novel, high-stakes questions that don't fit standard consulting frameworks.
We never lead with a model or a framework. We start by understanding the decision at hand — then choose the best tools from across our research disciplines to address it with precision.
Every engagement starts with a structured analysis of the choice being made — not the data, not the model.
We identify constraints, stakeholders, alternatives, and second-order effects before proposing any solution.
Only then do we reach into our toolkit — statistical models, ML systems, economic frameworks — matching technique to problem.
Solutions that are technically rigorous and intuitively explainable to any stakeholder who needs to act on them.
From supervised learning to reinforcement learning, we bring technical depth — and the judgment to identify where AI creates genuine leverage and where simpler methods work better.
Rigorous models capturing agent behavior, market dynamics, and incentive structures — using microeconomics, game theory, and mechanism design.
Precise formulations built from first principles — differential equations, optimization theory, stochastic processes, and more.
Applied research at the intersection of financial theory and modern ML — risk factor modeling, portfolio construction, systematic asset allocation, and market microstructure.
We draw on AI, ML, economics, and finance — because the best solution to your problem might live in any of them.
We never lead with a technology. We start with your decision problem and work backward to the right approach.
Complex solutions should be explainable. We build things that are technically sophisticated and humanly understandable.
Our consulting practice is grounded in active research. We stay at the frontier so our solutions reflect current best practice.
Whether you're an individual, a company, or a government agency — if you have data and a decision to improve, we'd like to hear about it.