The latest articles on DEV Community by PARIKSHIT SHARMA (@gamingboy). https://dev.to/gamingboy https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4078279%2F8732c773-cdfc-4ba9-a7bb-fadf0d4b3bac.png https://dev.to/gamingboy en PARIKSHIT SHARMA Fri, 14 Aug 2026 23:08:17 +0000 https://dev.to/gamingboy/i-built-a-portable-memory-layer-for-ai-agents-so-you-dont-have-to-3n8l https://dev.to/gamingboy/i-built-a-portable-memory-layer-for-ai-agents-so-you-dont-have-to-3n8l <p>markdown</p> <h1> I built a portable memory layer for AI agents so you don't have to </h1> <p>AI agents are everywhere now. But they still struggle with one thing: <strong>memory</strong>.</p> <p>Most agent frameworks store memory in a vector database, but that's not enough. A vector DB just stores embeddings; it doesn't understand memory types, consolidation, forgetting, or portability.</p> <p>So I built <strong>Mneme</strong> — a portable memory layer for AI agents.</p> <h2> What does Mneme do? </h2> <ul> <li> <strong>Structured memory</strong>: episodic (what happened), semantic (facts/preferences), procedural (how to behave).</li> <li> <strong>Semantic recall</strong>: find relevant memories using local embeddings (FastEmbed).</li> <li> <strong>Consolidation</strong>: deduplicate and summarise memories over time.</li> <li> <strong>Forgetting</strong>: delete memories with full audit trail.</li> <li> <strong>Portability</strong>: export/import your agent's entire memory to a <code>.mneme</code> file.</li> <li> <strong>Access control</strong>: multi‑agent scoping with explicit shared memory grants.</li> <li> <strong>Local‑first</strong>: SQLite backend, zero‑config.</li> </ul> <h2> Installation </h2> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code> bash pip install mneme-memory Usage python import mneme memory = mneme.Store(agent_id="my-agent", backend="memory.db") memory.remember("User prefers email over Slack", memory_type="semantic") context = memory.recall("How does the user like to be contacted?") print(context) That's it. Three verbs: remember, recall, forget. Why not just use a vector DB? A vector DB gives you similarity search, but not: Memory types (episodic vs semantic vs procedural) Consolidation (episodic → semantic summarization) Controlled forgetting + audit trail True portability across frameworks Mneme is not a database; it's a memory layer that can sit on top of SQLite or Postgres. Performance Retrieval precision@1: 1.00 on synthetic test Recall latency: 7.34 ms average Write latency: 0.088 ms Check it out GitHub: https://github.com/GamingBoyOfficial/Mneme PyPI: https://pypi.org/project/mneme-memory/ What's next? I'm planning to add HNSW vector index, TypeScript SDK, and more adapters. Let me know what you think! </code></pre> </div> ai opensource rust python