Reliability-first architecture for quantitative execution: deterministic state, resilient I/O, and operational safety primitives.
Continuous health monitoring for critical loops. Detects stalls, enforces timeouts, and triggers controlled recovery.
Write-then-swap persistence model to protect configuration and reporting artifacts against corruption and power loss.
Detects orphaned processes/state and restores a consistent runtime snapshot, minimizing downtime during reconnect events.
A health supervisor for critical loops (data feed, signal engine, order pipeline). It enforces heartbeat checks, detects stalls, and triggers controlled recovery with invariant validation.
For configuration, catalogs, and reports, Sentralyx uses a write-then-swap pattern. Critical artifacts are either fully written or unchanged—no partial states.
Disconnects happen. Orphan recovery reconciles open orders/positions after reconnect, restores a consistent snapshot, and prevents duplicate intent under uncertain network conditions.
Designed around least privilege and local-first secret handling. Use trade-only keys, IP restrictions, and encrypted storage. Logs and exports remain audit-friendly without leaking sensitive credentials.
Illustrative snippets showing how signals and confirmations are expressed.
A simple gate that favors directional regimes (ADX) and avoids overheated entries (RSI).
Cross events add structure to momentum shifts while keeping the rule expressive and inspectable.
Multi-timeframe confirmation reduces false positives by requiring alignment across regimes.
High-level reliability and quality posture indicators.