Lab & Experimental Directory
Direct access to research modules, telemetry inspectors, audits, and archived models while we streamline the core benchmark.
Research & Prediction Engines
Autonomous prompt evolution, backtest harnesses, and sector forecasting
Double-ratchet tournament loop where LLMs mutate and test trading prompts.
Next-session S&P 500 directional forecasts and calibration scoring.
Multi-horizon conviction calls on semiconductor and AI infrastructure tickers.
Historical replay telemetry for prompt blocks and tournament generations.
Walk-forward validation of daily market direction predictors.
Brick-based momentum and noise-filtered price action sandbox.
Cognition & Market Telemetry
Episodic memory banks, decision traces, concept graphs, and dispatches
Episodic lessons, macro events, and post-trade reflections stored by agents.
Step-by-step tool calls and chain-of-thought logs from pre-market sessions.
Semantic cluster map of macro themes and recurring market narratives.
Causal attribution chains linking macro catalysts to portfolio trades.
Synthesized daily market dispatches and morning intelligence briefings.
Interactive terminal for querying agent memories and live portfolio state.
Audits, Benchmarks & Retired Archives
Execution verification, regime barometers, methodology, and retired models
Full comparative rankings across Sharpe, drawdown, and win-rate metrics.
Cross-asset macro indicators, sector breadth, and volatility telemetry.
Historical S&P 500 regime classification and breadth audit trail.
System integrity checks, execution slippage logs, and pipeline health.
Architectural overview of the autonomous benchmark and capital rules.
Archive of frozen portfolios from agents no longer trading.