用 NVIDIA NeMo Agent Toolkit 与 Amazon S3 Vectors 构建智能体记忆
原标题：Build agent memory with NVIDIA NeMo Agent Toolkit and Amazon S3 Vectors
AI 摘要
AWS 机器学习博客发布教程，介绍如何将 Amazon S3 Vectors 作为持久化记忆层集成到 NVIDIA NeMo Agent Toolkit（NAT）中，并部署在 Amazon EKS 上。文章以多智能体投资研究为示例，讲解 NAT 记忆子系统（MemoryEditor、MemoryItem 等）的工作方式，以及如何实现自定义记忆提供者。该方案面向需要弹性向量存储、强写一致性和低成本扩展的生产级多智能体系统。
正文节选
Build agent memory with NVIDIA NeMo Agent Toolkit and Amazon S3 Vectors In my previous post: Building persistent memory for multi-agent AI systems with Amazon S3 Vectors, we explored why memory engineering is the foundational discipline for production multi-agent systems. We showed how Amazon S3 Vectors, a capability of Amazon Simple Storage Service (Amazon S3), meets the architectural requirements for agent memory: semantic retrieval, rich metadata, strong consistency, and elastic scale. In thi