Orris

Create an image for this

Reading File
Generating Image
OR Assistant Architecture
Here's the architecture diagram covering all the key components from your document:
  • Phase 1 (Pre-OT) on the left showing the ingestion pipeline: documents → Unsiloed parser → chunk/embed → MOSS vector DB
  • Phase 2 (Intra-OT) on the right with the two parallel planes:
    • Logging Plane (always-on, async): LiveKit → STT → Logger loop
    • Query Plane (on-demand): voice trigger → context assembly → MOSS search → LLM → MiniMax TTS + React UI
  • Guardrails at the bottom (source attribution, confidence threshold, hallucination guard, sterility guard)
Let me know if you'd like any changes to the layout, color scheme, or level of detail.
This is a shared conversation. Sign in to Orris to start your own chat.