Every trade teaches the system what should have happened.
Loch Raven Capital is our own fund, run end to end by AI agents and graded against the S&P 500 every trading day. It exists to prove one thing in public: systems can learn. The same loop we run on trades, we install on the decisions your organization repeats.
Paper portfolio the agents manage end to end, launched June 2026
Of the S&P 500 since launch, graded every trading day
Decisions journaled, graded, and fed back as lessons
In plain terms: the desk makes a call, writes down why, checks the next day what actually happened, and adjusts. Four strategy sleeves run research, sizing, and execution inside human-set guardrails: risk caps, circuit breakers, and a decision journal we can audit line by line. It runs as a paper desk by design, so every decision is auditable before a dollar ever moves.
Continuous Decision Architecture, live
The fund is the public proof of our methodology. The same four-stage loop runs on any decision your organization repeats: pricing, inventory, budgets, campaigns.
Decide
The system makes or drafts the call inside guardrails you set: risk caps, circuit breakers, human-visible journals.
Record
Every decision is journaled with the reasoning behind it, not just the outcome. Nothing important lives in someone’s memory.
Grade
Each call is scored against what actually happened, including the counterfactual: what would have happened if you had passed.
Adjust
Thresholds tune themselves from the scorecard. The system becomes more careful exactly where it has been wrong.
The whole practice, one page
The fund is one exhibit among many. The same engineering shows up in the Morgan State portal, the Urban League tracker, and every system in the gallery, and this overview puts them side by side.
The fund as a client systemLoch Raven Capital is our own research desk running a simulated portfolio. Nothing on this page is investment advice or an offer to invest. Performance figures are measured by the desk against the S&P 500 and do not predict future results. The desk appears here to demonstrate engineering capability.
