Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Governed Memory: A Production Architecture for Multi-Agent Workflows

Hamed Taheri

Abstract

Enterprise AI deploys dozens of autonomous agent nodes across workflows, each acting on the same entities with no shared memory and no common governance. We identify five structural challenges arising from this memory governance gap: memory silos across agent workflows; governance fragmentation across teams and tools; unstructured memories unusable by downstream systems; redundant context delivery in autonomous multi-step executions; and silent quality degradation without feedback loops. We present Governed Memory, a shared memory and governance layer addressing this gap through four mechanisms: a dual memory model combining open-set atomic facts with schema-enforced typed properties; tiered governance routing with progressive context delivery; reflection-bounded retrieval with entity-scoped isolation; and a closed-loop schema lifecycle with AI-assisted authoring and automated per-property refinement. We validate each mechanism through controlled experiments (N=250, five content types): 99.6% fact recall with complementary dual-modality coverage; 92% governance routing precision; 50% token reduction from progressive delivery; zero cross-entity leakage across 500 adversarial queries; 100% adversarial governance compliance; and output quality saturation at approximately seven governed memories per entity. On the LoCoMo benchmark, the architecture achieves 74.8% overall accuracy, confirming that governance and schema enforcement impose no retrieval quality penalty. The system is in production at Personize.ai.

Governed Memory: A Production Architecture for Multi-Agent Workflows

Abstract

Enterprise AI deploys dozens of autonomous agent nodes across workflows, each acting on the same entities with no shared memory and no common governance. We identify five structural challenges arising from this memory governance gap: memory silos across agent workflows; governance fragmentation across teams and tools; unstructured memories unusable by downstream systems; redundant context delivery in autonomous multi-step executions; and silent quality degradation without feedback loops. We present Governed Memory, a shared memory and governance layer addressing this gap through four mechanisms: a dual memory model combining open-set atomic facts with schema-enforced typed properties; tiered governance routing with progressive context delivery; reflection-bounded retrieval with entity-scoped isolation; and a closed-loop schema lifecycle with AI-assisted authoring and automated per-property refinement. We validate each mechanism through controlled experiments (N=250, five content types): 99.6% fact recall with complementary dual-modality coverage; 92% governance routing precision; 50% token reduction from progressive delivery; zero cross-entity leakage across 500 adversarial queries; 100% adversarial governance compliance; and output quality saturation at approximately seven governed memories per entity. On the LoCoMo benchmark, the architecture achieves 74.8% overall accuracy, confirming that governance and schema enforcement impose no retrieval quality penalty. The system is in production at Personize.ai.
Paper Structure (54 sections, 4 figures, 8 tables)

This paper contains 54 sections, 4 figures, 8 tables.

Table of Contents

  1. Introduction
  2. The Memory Governance Gap
  3. Why RAG Is Not Enough
  4. Contributions
  5. Related Work
  6. Retrieval-Augmented Generation and Iterative Retrieval
  7. Agent and Production Memory Systems
  8. Evaluation and Context Delivery
  9. Positioning Summary
  10. Architecture Overview
  11. Dual Memory Model
  12. Problem Statement
  13. Open-Set Memory
  14. Schema-Enforced Memory
  15. Algorithm: Dual Extraction Pipeline
  16. ...and 39 more sections

Figures (4)

  • Figure 1: Governed Memory four-layer architecture. Agent nodes interact through a shared, organization-scoped API surface. Layers are independently configurable.
  • Figure 2: Dual extraction pipeline. A single LLM call produces both open-set facts and schema-enforced typed properties, followed by quality gates and write-side deduplication.
  • Figure 3: Governance routing with tiered modes (fast/full) and progressive context delivery. The session layer tracks delivered variables to inject only delta content on each autonomous step.
  • Figure 4: Reflection-bounded retrieval with entity-scoped isolation. The optional reflection loop generates targeted follow-up queries within bounded rounds; entity isolation is enforced by CRM key pre-filtering.