The Modern Revolution Fund LP
Fourth Turning Partners is the general partner of The Modern Revolution Fund — a highly liquid, transparent fund trading digital assets, options, futures, and equities, guided by proprietary AI research and disciplined risk management.
Accredited investors only.
Thesis
Strauss and Howe described history as a recurring cycle of four generational turnings, with the fourth — the crisis — as the period when a society's core institutions are remade. We are living through one of those periods now.
Money is one of those institutions. The first financial revolution was double-entry bookkeeping. The second was the joint-stock company. The third was the modern central bank. The fourth is the convergence of programmable money, artificial intelligence, and globally accessible markets — and it is happening now.
Fourth Turning Partners exists to invest through this transition. We treat AI not as a marketing label but as core research and implementation infrastructure. We treat risk management as the central discipline, not an afterthought. And we treat transparency as a feature, not a compliance burden.
The institutions of the last era do not survive the fourth turning intact. The financial system is one of them.
Strategy
We build and operate AI systems for signal discovery, execution, and portfolio construction. AI is research and engineering, not narrative.
We invest across digital assets, options, futures, and equities. The fund is open-ended and liquid by design.
Position sizing, drawdown control, and explicit scenario analysis are the spine of the strategy — not an overlay.
Capital allocation to autonomous systems demands rigorous evaluation, monitoring, and override. We treat safety as part of the alpha.
The Modern Revolution Fund LP is a Delaware limited partnership and the investment vehicle for the strategy. The fund holds a highly liquid book with a 90-day initial lockup and a 14-day withdrawal period thereafter. Interests are offered under Rule 506(c) of Regulation D to verified accredited investors only.
Approach
Limited partners see positions, methodology, and exposures. Black-box marketing is not a strategy and it is not an edge.
The fund trades liquid instruments and accepts withdrawals on a 14-day notice period after a 90-day initial lockup. Liquidity terms should match what investors actually need, not what is convenient for the manager.
Meaningful general-partner capital is committed alongside limited partners. The principals invest in the same fund, on the same terms.
Principals
Co-Founder & Chief Investment Officer
Leads firm strategy. Teaches generative AI at Vanderbilt University, where he is on the faculty of the School of Engineering. Background spans AI research, large-scale software engineering, and applied risk modeling across digital and traditional markets. Author of the SIEVE framework for algorithmic strategy evaluation and the Mangrove Memory architecture for agent reasoning.
Co-Founder & Chief Executive Officer
Leads investor relations and operations. Owns capital formation, LP communications, fund administration, regulatory engagement, and the firm’s relationships with counsel, custodians, and service providers. Responsible for the operational discipline that underwrites the investment process and the long-term institutional posture of Fourth Turning Partners.
Research
Working paper · Under review
Timothy Darrah · Mangrove Technologies / Vanderbilt University
SIEVE (Strategy Indicator Embeddings with Value Estimation) is the learned representation of technical indicators that produces this prediction. Trained on 1.1 million backtests across 63 assets and 136 indicators, it achieves 0.976 AUC on the trade/no-trade task and identifies 83.5% of historical strong-winner strategies. Deployed as a pre-filter inside our research pipeline, SIEVE has cut backtesting compute by 64.3% across 1.24 million production runs while retaining approximately 95% of winning strategies.
Preprint forthcoming on this site.
Working paper · Under review
Timothy S. Darrah · Vanderbilt University
Mangrove Memory is the architecture that delivers this. It is grounded in the cognitive science of human memory rather than the engineering convenience of vector stores: a formal ontology drawn from Knowledge Space Theory sits as middleware in the conversation loop, the agent never operates the memory system directly, and memory surfaces through automatic activation rather than deliberate query. Consolidation runs offline between sessions, on the same logic that makes biological memory consolidate during sleep.
Preprint forthcoming on this site.
Contact
We welcome inquiries from accredited investors and their advisors. We typically respond within two business days.
After we confirm fit, we send the Confidential Private Placement Memorandum, Limited Partnership Agreement, and Subscription Documents under separate cover. Subscriptions are accepted only after formal verification of accredited-investor status, in accordance with Rule 506(c).