JD Health rebuilt a four-layer engineering stack — model, knowledge, harness, and delivery — replacing probabilistic RAG with deterministic Agentic Search and structured knowledge base LLMwiki to achieve verifiable, auditable medical AI across clinical tasks.
JD Health rebuilt its medical AI stack across four layers, replacing probabilistic RAG with deterministic Agentic Search and structured knowledge base LLMwiki to achieve verifiable, auditable clinical task execution.
ByteDance's Volcano Engine open-sourced OpenViking, a self-evolving context database for AI agents that replaces vector stores with a viking:// virtual file system using three-layer progressive loading (L0/L1/L2), hierarchical retrieval with visible traces, and automatic long-term memory extraction, cutting input tokens 34-91% and boosting LoCoMo benchmark scores while integrating with Claude Code, Codex, and other tools.
OpenViking, an open-source context database from Volcano Engine with 32k GitHub stars, solves AI agent memory overflow by storing long-term memory, project resources, and skills in a virtual file system with three-layer summaries, traceable retrieval, and automatic memory consolidation, integrating with major coding agents like Codex and Cursor.
This article explains how to create a two‑layer AI knowledge system— a dynamic Knowledge Base Layer and a static Brand Foundation Layer— in about 20 minutes, detailing its architecture, advantages over traditional RAG, step‑by‑step deployment, and real‑world use cases for creators, teams, and personal productivity.
Karpathy’s LLM Wiki proposes a meta‑framework that lets large language models continuously compile, update, and query a structured Markdown wiki, moving beyond traditional RAG by treating ideas as reusable assets that agents can automatically materialize into personal knowledge bases.
