PayQIA
Back to Products
Core Primitive 01

Memory Graph Layer

A dynamic knowledge graph connecting conversational facts, user preferences, and temporal relationships.

Entity Resolution

When an agent ingests information (e.g. "I love React" and "My favorite frontend tool is React"), the QIA engine automatically resolves synonyms, links entities, and groups them under singular conceptual nodes.

Temporal Semantic Decay

Memories fade or strengthen dynamically. Nodes that aren't accessed decay over time, while frequently retrieved nodes form stronger conceptual pathways, ensuring only high-signal contexts are recalled.

Clustering & Resolving Graph States

graph-resolve.ts
import PayQIA from "payqia";

const client = new PayQIA({ apiKey: process.env.PAYQIA_API_KEY });

// Ingest chat log facts and automatically trigger entity resolution
const graphState = await client.graph.resolve({
  containerTag: "user_session_4821",
  decayThreshold: 0.15, // Low scoring stale memories prune automatically
  resolveSynonyms: true
});

console.log(graphState.nodes); // e.g. [{ entity: "React", strength: 0.98 }]
Theoretical Foundations

Read the Entity Resolution & Decay Research

Explore the theoretical math, cognitive models, and shared memory swarm graphs on the Haqikos Research portal.

/research/haqikos