Your AI's memory layer
Igloo builds a living knowledge graph from your email, calendar, meeting transcripts, and documents — so your AI already has the context it needs.
Works with Claude, ChatGPT, Cursor, and any MCP-compatible tool
The difference
Since Meridian Health is a mid-market health system like NovaCare, I'm basing this on the NovaCare proposal from Feb 14 and the BrightPath proposal from Jan 22 — both closed in the healthcare vertical.
A few key differences to account for:
Drafting now using NovaCare structure with these adjustments...
Your knowledge graph
Igloo turns raw data into structured, queryable knowledge — people, decisions, topics, and meetings your AI tools can actually use. This is a real, interactive graph. Try it.
Connect Gmail or Google Calendar and watch your graph build itself.
Your data
We believe you should know exactly what's stored and what isn't.
Raw data
Email bodies and calendar details are processed but never stored
Structured references
Only metadata — subject lines, dates, senders, and participants
Knowledge graph
People, decisions, topics — connected and queryable
Encrypted at rest. Raw content never retained.
Your data is encrypted using industry-standard encryption. Igloo processes raw content to build your knowledge graph, then discards the source material. Only structured references remain.
Integrations
Your AI tools don't just read. They write back.
Decisions, context, memory — your graph grows from use.