Minutes, Not Weeks: Long-Term Memory for AI Applications With PowerMem
Talkat KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026 · 2026-09-07..09-09·
summary
PowerMem is an open-source Apache 2.0 memory library for AI applications that provides a single memory API accessible from five surfaces: a Python SDK, a pmem CLI, an HTTP API (with OpenAPI docs), an MCP server (SSE/stdio/streamable-HTTP), and a web dashboard. It focuses on developer experience with features like auto-loading .env config for minimal Python examples, pluggable storage backends (OceanBase, PostgreSQL, SQLite) and model providers so prototypes and production share code paths, plus built-in CLI tools for schema migration and backup/restore.
AI-written summary · about
related talks
46mThe Anatomy of Memory in Humans and AI Agents
27mMemory Engineering: Going Beyond Context Engineering
54mThe Anatomy of Memory in Humans and AI Agents
01h 49mBuilding Memory Augmented Agents with MemoRizz
21mBuilding memory for AI Agents
10mRapid Prototyping in Next.js: Building a Viral Ai Tool in 8 Hours
15mMemory-Augmented AI Agents: Enhancing Long-Term Reasoning and Context Awareness
03mKeynote: Open Models Need Open Memory