Welcome to the
Help Center.

Evnao is an enterprise-grade AI customer support platform. It replaces legacy, decision-tree chatbots with a semantic AI engine that can understand complex customer intents, fetch data from your knowledge base, and even render interactive UI components (Generative UI) directly in the chat to resolve issues instantly.

Never. We enforce a strict Zero Data Retention (ZDR) policy with all our foundation model providers (OpenAI, Anthropic). Your data is processed ephemerally to generate a response and is instantly discarded. It is never used to train public models.

Evnao detects the customer's language automatically on their first message and responds flawlessly in over 50 languages, including complex right-to-left languages like Arabic and Hebrew. You do not need to translate your knowledge base; the AI reads your English docs and translates on the fly.

Yes. Enterprise customers can bypass our internal routing and use their own API keys for OpenAI, Anthropic, or even custom fine-tuned models hosted on AWS SageMaker or Azure.

Evnao provides native, one-click integrations with Zendesk Guide, Intercom Articles, and Notion. We sync your data in real-time. If you update a return policy in Zendesk, the Evnao AI knows about it within 60 seconds.

You can upload PDF, DOCX, TXT, MD (Markdown), and HTML files directly via the Dashboard or the REST API. Max file size is 50MB per document.

Evnao uses an advanced RAG (Retrieval-Augmented Generation) pipeline. By default, the AI is strictly instructed to only answer questions using the context provided in your Knowledge Base. If the answer doesn't exist, it is programmed to say 'I don't know' and route the conversation to a human.

Yes. By passing custom metadata (like a user ID or Stripe Customer ID) when initializing the widget, the AI can look up live account statuses, active subscriptions, and recent orders without asking the user to authenticate twice.

Human agents use the Evnao 'Team Inbox'. When the AI routes a ticket, the agent sees the full chat history, a concise AI-generated summary of the problem, and the customer's intent. The agent takes over seamlessly.

Yes, but safely. Through our Generative UI system, Evnao can generate a refund UI card. You can set Routing Rules so that for actions over a certain dollar amount, the AI must ping a human 'Reviewer' for a one-click approval before the action executes.

Handoffs occur automatically if: 1) The user explicitly asks for a human. 2) The AI detects high negative sentiment. 3) The AI cannot find the answer in the Knowledge Base. 4) A custom Routing Rule you defined is triggered.

Absolutely. You can create rules like: 'If intent = billing, route to Finance Team' or 'If sentiment = angry, route to Escalations Team'.

If you are using React, Next.js, or standard HTML, it takes less than 5 minutes. You simply drop the ChatWidget component or our `script` tag into your application.

Yes. We fire real-time webhooks for events like `conversation.started`, `conversation.ended`, and `handoff.requested`. All webhooks are secured with an HMAC SHA-256 signature.

Standard plans are limited to 1,000 requests per minute. Enterprise plans have custom rate limits and dedicated infrastructure to ensure no throttling during high-volume events like Black Friday.

Evnao is SOC 2 Type II certified, GDPR compliant, and HIPAA compliant. All data is encrypted at rest (AES-256) and in transit (TLS 1.3). We also automatically redact PII (SSNs, credit cards) before storing logs.