Notes, documents, quotes, embeds, files, saved AI chats — every neuron is auto-described, auto-tagged, and embedded in your language. One memory. Every AI you use plugs in.
Color-coded neurons — notes purple, documents blue, quotes yellow, embeds teal.
"That thing about the Berlin trip" finds the note titled "European roadmap — Q3" because they're about the same topic.
Save in Dutch. Ask in English. Cortex embeds in the source language and matches across — no translation step, no missed hits.
Every neuron tracks its history with diff view and one-click restore. A private working_notes scratchpad keeps drafts separate from what you share.
Set the active project; Claude, ChatGPT, Cursor all default to that scope within a second. Switch scope from any client and everyone follows.
Type, paste, drop a file, forward an email, save a chat from Claude Desktop — every path creates a neuron.
Cortex generates embeddings, descriptions, and tags in seconds. Hybrid retrieval: dense + sparse + reranked.
Search from the brain view, ask in chat, or call from Claude Code, Cursor, and Claude Desktop via MCP.
Still curious? Write to us.
Private beta. Limited spots. Redeem your code to jump in — or join the waitlist at the bottom of the page.