Thoughts on AI memory, persistent context, and autonomous development.
Why AI Memory Changes Everything
The missing infrastructure layer between AI assistants and AI agents.
2026-03-11
AI Memory vs RAG vs Fine-Tuning
Three approaches to persistent AI context. One clear winner for coding assistants.
2026-03-11
Memento vs Claude-Mem vs Mem0 vs Supermemory vs Context Cloud vs Zep: AI Memory Comparison (2026)
Memento vs Claude-Mem vs Mem0 vs Supermemory vs Context Cloud vs Zep: honest comparison of AI memory solutions for coding assistants.
2026-09-25
Shared Cursor Memory for Teams: Hosted MCP Memory (2026)
Cursor forgets everything between threads. Memento gives your whole team one shared, hosted memory over MCP: markdown pages every teammate's agent reads, writes, and recalls.
2026-09-25
Cursor forgets the next chat: add hosted MCP memory (2026)
A new Cursor Agent chat does not have yesterday's repo conventions. Connect hosted Memento over MCP so the next session can recall.
2026-09-26
Claude Code forgets the next session: add hosted MCP memory (2026)
A new Claude Code session does not have last night's repo notes. Connect hosted Memento over MCP so the next session can recall.
2026-09-26
Codex persistent memory over MCP (2026)
A new Codex or ChatGPT coding run does not have last session's repo notes. Point the client at hosted Memento over MCP so the next run can recall.
2026-09-26