✕ Close
Home Mandate Engine Contact
Σ
Systematic Alpha · Autonomously Discovered

Markets yield to those who reason without fatigue.

Arctix Capital deploys AI-driven strategy research at continuous scale — discovering, validating, and executing systematic edge across equity markets.

1.5+ Sharpe threshold
14 Day validation min.
24/7 Autonomous ops
Strategy Generation Backtesting Evaluation Promotion Paper Trading Validation Live Deployment Memory Update Strategy Generation Backtesting Evaluation Promotion Paper Trading Validation Live Deployment Memory Update

Edge discovered by machines, validated by rigour.

Traditional quantitative research is bottlenecked by human throughput. Arctix operates differently — our AI engine generates and stress-tests hundreds of strategies in parallel, promoting only those that clear a demanding statistical threshold.

The result is a continuously evolving portfolio of systematic strategies, each with a documented edge and a clear reason to exist in markets. No discretion. No override. No fatigue.

01
Autonomous Generation

Our AI engine produces novel strategies around the clock — exploring signal combinations, timeframes, and market regimes without human prompting or bias.

02
Rigorous Validation

Every candidate clears Sharpe ≥ 1.5, max drawdown ≤ 15%, and a minimum 14-day live paper trading period before any capital is considered.

03
Adaptive Memory

The system learns from each cycle — exploiting proven approaches while systematically exploring adjacent strategy space at a 70/30 exploitation-to-exploration ratio.

Five components.
One continuous loop.

From raw idea to validated live strategy — the engine runs without pause, without instruction, without limit.

Step 01
Generate
AI synthesizes novel strategy logic
Step 02
Backtest
Vectorbt simulation across history
Step 03
Evaluate
Sharpe, drawdown, win rate gates
Step 04
Validate
14-day paper trading proof
Step 05
Deploy
Live Alpaca execution
GEN — 01
Strategy Generator

Claude Sonnet synthesizes novel trading logic from microstructure signals, factor research, and pattern libraries — producing executable specifications at continuous scale.

MEM — 02
Memory System

Persistent strategy memory informs each generation cycle. What works is remembered. What fails is learned from. The 70/30 exploitation-exploration ratio governs each batch.

ORC — 03
Orchestrator

The central coordinator runs continuous batch cycles — pacing API calls, routing strategies through the pipeline, and maintaining the system's operational tempo.

Discipline over discretion. Always.

Performance reported on paper-trading basis during engine validation phase. All figures reflect systematic execution without manual override. No exceptions.