ASMO Intelligence Engine

System Architecture

Real backend infrastructure — RSS ingestion → 9-agent AI pipeline → personalization → learning loop. Every node reflects actual system state.

Architecture

Ingestion Layer

RSS SourcesIntelligence feeds
Scraperfeedparser + ArticleService
Offline
Article StorePostgreSQL — articles table
Intelligence Pipeline — LangGraph DAG
Event AnalysisEventAnalysisAgent
Market PredictionsMarketPredictionAgent
Confidence ReviewConfidenceReviewAgent
Intelligence DecisionIntelligenceDecisionAgent
Scenario IntelligenceScenarioIntelligenceAgent
Alert EngineAlertAgent
ReflectionReflectionAgent
MemoryMemoryAgent + intelligence_memory

Personalization Layer

Alert Candidatesalert_candidates table
Personalization EngineAlertPersonalizationService
User AlertsMarket + equity matching

Learning Loop

Agents & Engines

ASMO is not a single LLM call. Each stage is handled by a specialized agent with defined inputs, outputs, and responsibilities.

Frontend Connections