All memos tagged #context-management

MemPalace, Obsidian, and MythOS are three different answers to the same question: how does an AI collaborator carry knowledge across sessions? They sit at three distinct points on the spectrum from...

Progressive disclosure is the practice of keeping an AI agent's baseline context minimal and loading instructions, tools, and knowledge only at the moment a task actually calls for them. The term...

The new rules of context engineering for Claude 5 generation models are six then/now shifts moving agent Context Engineering away from prescriptive scaffolding and toward judgment, interface design,...

CLAUDE.md is Claude Code's project-level instruction file — a Markdown document that loads into every session's context window and steers agent behavior for that project. It lives at the root of a...

MemPalace and Obsidian are two of the most common ways to give Claude persistent memory. They solve different halves of the problem and are, in practice, complementary rather than competing....

"What's the best memory system for Claude?" is one of the most-asked questions in AI-builder communities — and the three most common answers are MemPalace, Obsidian, and MythOS. Each solves a...

People using Obsidian as a memory system for Claude are building something MythOS was designed to be from day one. Here's what they're doing, why it works, and where MythOS is the better architecture...

On April 2, 2026, Andrej Karpathy posted a detailed architecture for "LLM Knowledge Bases" — personal knowledge systems built and maintained by LLMs, stored as markdown, compounding with use. He...

Token Optimization is the practice of managing language-model token usage efficiently across agent systems — reducing operating costs in long-running, multi-agent, or autonomous workflows without...

On April 3, 2026, Andrej Karpathy posted a detailed architecture description of "LLM Knowledge Bases" that landed as one of the most significant public signals in AI-native knowledge infrastructure....

Context engineering is the practice of deliberately shaping the information environment an AI agent observes at every step—its system instructions, retrieved knowledge, and conversational memory. AI...

MemPalace and MythOS solve adjacent problems with fundamentally different architectures. MemPalace is an AI memory backend — it makes agents remember. MythOS is a knowledge platform — it makes...

A practical guide to preventing CLAUDE.md bloat by treating it as a steering document synced to an external knowledge system. The Problem Claude Code's CLAUDE.md file loads into every session's...

GBrain is an open-source Personal Knowledge Management system designed for AI agents, created by Garry Tan, president and CEO of Y Combinator. Built in TypeScript on Bun, it transforms meetings,...

AI memory systems are the tools and architectures that give Artificial Intelligence models persistent context across sessions — the ability to remember who you are, what you've worked on, and what...