How it works · Architecture & data flow

Transparent by design — see exactly what happens

What happens when someone talks to a daita agent, the technology behind each step, and where your data lives. No black box.

The data flow

From a question to a completed task

A streaming agentic pipeline — grounded in your knowledge and catalog, bounded by guardrails, observable end to end.

  1. 1
    A visitor reaches outWeb chat, voice, phone or WhatsApp — one agent with shared memory across every channel.
  2. 2
    Verified, encrypted ingressThe request enters daita over encrypted channels; phone & WhatsApp webhooks are cryptographically verified before anything runs.
  3. 3
    Claude reasons and plansAnthropic Claude reasons through the task in a streaming agentic loop and decides which knowledge and tools it needs.
  4. 4
    Grounds in your knowledge and catalogTwo-stage retrieval finds candidates, then a neural reranker (EU) orders them by true relevance. Answers are cited, never invented.
  5. 5
    Acts with real toolsSearches your catalog, builds carts, checks orders, books calendar slots, captures leads and updates your CRM — real actions, not just talk.
  6. 6
    Responds in text or neural voiceReplies in text and natural neural voice by default — or a live speech-to-speech call if you opt in — and can even navigate the page for the visitor.
  7. 7
    Records to an EU database, escalates when neededEvery interaction is written to an EU database with an immutable audit log; when human judgement is needed, it hands off with the full conversation.
Your data at rest lives in the European Union · audit logs are immutable · full EU-region or on-prem deployment available
The technology

The stack inside, by layer

Best-in-class components, each with a clear role — and a clear data-residency posture.

EU-region / on-prem available

Anthropic Claude

Reasoning, language understanding, tool use (agentic loop)

EU

BM25 + Jina Reranker (Berlin)

Two-stage retrieval over your knowledge, with citations

EU-region / on-prem available

ElevenLabs neural voice

Natural spoken replies — the default voice

Opt-in · may leave the EU

OpenAI gpt-realtime

Live speech-to-speech voice — optional, owner-enabled

EU-region / on-prem available

Twilio · Web · Shopify

Phone, WhatsApp, embeddable web widget & Shopify app

EU

Firestore (eur3, Europe)

Leads, bookings, transcripts and immutable audit logs

EU-region / on-prem available

Resend (SPF/DKIM/DMARC)

Branded, fully deliverable confirmations and calendar invites

EU

Self-hosted observability

Latency, tool use and errors — your data, no third-party analytics

Where your data lives

Where your data lives

daita is EU-first. Data at rest is stored in the EU by default (Firestore eur3) and semantic search runs in the EU (europe-west1). Some AI processing — the language model and opt-in live voice — currently runs via vetted subprocessors under EU-approved safeguards (DPF/SCCs), and is migrating to Claude on Vertex AI in the EU. Anything that can leave the EU is opt-in and clearly flagged.

Data at rest in the EU by default (Firestore eur3)
Semantic search in the EU (europe-west1)
Some AI processing via vetted subprocessors under EU safeguards (DPF/SCCs) — migrating to Claude on Vertex AI in the EU
GDPR & EU AI Act aligned; immutable audit logs
daita · München

Want this for your business?

Talk to a live agent on this site right now, or get your own in minutes.