#
Xentropy's AI chat is grounded in your knowledge — it doesn't just generate answers from its training data, it searches your notes, voice memos, and captures to give you answers that are yours.
Status: AI chat with conversation history exists. Retrieval-augmented generation (RAG) over your personal knowledge is being wired up and will ship soon.
#
The AI chat uses a multi-agent architecture with specialized roles:
| Agent | Role | |-------|------| | Router | Understands your intent and routes to the right agent | | Researcher | Searches your knowledge graph for relevant notes | | Writer | Drafts responses in a natural, conversational tone | | Critic | Reviews answers for accuracy and flags uncertainty |
This means when you ask "What did I decide about pricing?", the AI:
- Interprets the question as a knowledge-retrieval request
- Searches your notes for decisions related to pricing
- Synthesizes a grounded answer with citations to source notes
- Flags confidence level and any contradictions found
#
Beyond the standard Q&A, Xentropy supports lens-based conversations where you can invoke different agent personas:
- Researcher lens — "Research the pros and cons of a React Native rewrite"
- Planner lens — "Help me plan the Q4 product roadmap"
- Devil's advocate lens — "Push back on this strategy — what am I missing?"
Switch lenses mid-conversation to get different perspectives on the same topic.
#
Unlike "chat with your notes" features in other tools:
- Not a sidebar chatbot — AI is woven into the surface, not bolted on as a panel
- Grounded in your graph — answers cite specific notes you can verify
- Multi-perspective — switch lenses to change how the AI approaches your question
- Works across all content — search voice memos, text notes, and photos in one query
#
- Retrieval-Augmented Generation (RAG) — wire your knowledge graph into every chat response with source citations
- Proactive briefings — daily "your mind this week" summaries
- Open-loop tracking — the AI reminds you of unresolved questions and promises
- Personal model — the AI learns your voice, recurring themes, and goals