---
type: "post"
id: "memory-as-middleware-for-self-improving-ai-agents-abd8b2"
slug: "memory-as-middleware-for-self-improving-ai-agents-abd8b2"
title: "Memory as Middleware for Self-Improving AI Agents"
description: "A paper submitted to arXiv on September 25, 2026 argues that durable memory for AI agents should be implemented as pluggable middleware rather than hand-wired into individual agents."
retrieval_nugget: "A paper submitted to arXiv on September 25, 2026 argues that durable memory for AI agents should be implemented as pluggable middleware rather than hand-wired into individual agents."
published_at: "2026-10-23"
updated_at: "2026-10-23"
record_date: "2026-10-23"
date_kind: "published_at"
topics: ["claude","claude-code","codex"]
entities: ["ALTK-Evolve","AppWorld","GPT-4.1","Claude Code","Codex","IBM Bob","pgvector","Milvus"]
related_patterns: ["company-memory-needs-write-loops"]
novelty_label: "new_fact"
what_changed: "A new research paper frames agent memory as pluggable middleware and presents ALTK-Evolve as a reference implementation, outlining six systems challenges beyond prior studies of memory performance."
prior_refs: ["memory:5195701c-913b-888d-a233-ff190763f6df","memory:79ef826c-5556-874b-aa0c-c90d3b6ff4b2"]
source_url: "https://arxiv.org/abs/2609.32091"
source_urls: ["https://arxiv.org/abs/2609.32091"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:memory-as-middleware-for-self-improving-ai-agents-abd8b2"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/memory-as-middleware-for-self-improving-ai-agents-abd8b2/","markdown":"https://newruntime.com/posts/memory-as-middleware-for-self-improving-ai-agents-abd8b2.md","json":"https://newruntime.com/posts/memory-as-middleware-for-self-improving-ai-agents-abd8b2.json"}
---

# Memory as Middleware for Self-Improving AI Agents

## Retrieval answer

A paper submitted to arXiv on September 25, 2026 argues that durable memory for AI agents should be implemented as pluggable middleware rather than hand-wired into individual agents.

A paper submitted to arXiv on September 25, 2026 argues that durable memory for AI agents should be implemented as pluggable middleware rather than hand-wired into individual agents. The authors describe ALTK-Evolve as a reference implementation for self-improving agents.

The paper identifies six design challenges: two-sided pluggability, host-native interposition, multi-tenant isolation, write-path consistency, federated sharing with provenance, and lifecycle governance. It lists host bindings for Claude Code, Codex, Claw Code, and IBM Bob, plus file, pgvector, and Milvus storage backends. Its agent-facing retrieval can return an always-on core of high-support guidelines alongside task-relevant results.

In AppWorld Test-Normal experiments using a GPT-4.1 agent, the authors report task-goal completion increasing from 69.6% to 73.2%, and scenario-goal completion from 50.0% to 64.3%. On the Hard tier, they report scenario-goal completion rising from 19.1% to 47.6%.

The source is a conditionally accepted research paper, not a product announcement. It does not establish independent replication, production reliability, availability of the listed integrations, or that the reported results generalize beyond its evaluated setup.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Claude](https://newruntime.com/topics/claude.md): Explore the claude topic hub.
- topic hub: [Claude Code](https://newruntime.com/topics/claude-code.md): Explore the claude-code topic hub.
- topic hub: [Codex](https://newruntime.com/topics/codex.md): Explore the codex topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)