AI Assistants · Powered by Emly

AI assistants that answer from your content, in your own cloud.

Datoin deploys Emly, an open-source AI assistant platform, inside your infrastructure. Every answer cites the page it came from, you use your own model key, and you're live in 14 days.

Send us your website URL. We'll build a working assistant on your public content, free.

Support assistanton your website

Can I export my reports to CSV on the Starter plan?

Assistant

Yes. Open any report and choose Export → CSV. Starter plans can export up to 10,000 rows per report[1].

[1] Help center · Exporting reports

Two Ways to Start

Pick the package that matches who the assistant is for. Both are deployed in your cloud, fixed-price, and live in 14 days.

For SaaS support teams

Launch in 14

Let your docs answer the repeat tickets. Live on your site in 14 days.

A customer-facing assistant that answers from your help center and docs, and hands buyers to sales when they ask about pricing.

  • Up to three assistants, such as support, docs and pricing
  • Your branding on the website widget or inside your product
  • One channel: website, in-product, Slack, Teams, Google Chat, WhatsApp or Telegram
  • Training and 30 days of support after go-live

For security-conscious organisations

Private AI Workspace

The AI assistant your security team will approve.

An assistant for each department in Slack, Teams, Google Chat, WhatsApp, Telegram or your intranet, answering from internal policies, handbooks and runbooks.

  • One assistant per department, each with its own documents and admins
  • Company sign-in through your identity provider, with role-based access
  • Citations and an audit log so answers and changes can be checked
  • A security pack with architecture and data flows for your reviewers

See It Working Before You Commit

Every engagement starts with proof on your own content, then a fixed-price package, then optional ongoing care.

Free Demo in 48 Hours

Send your website URL. We send back a working assistant built on your public content.

Scope and Fixed Price

A 30-minute call to agree the first assistants, channel and success measures.

Live in 14 Days

Deployed in your cloud, connected to your content and tested on real questions.

Optional Care Plan

A monthly review of unanswered questions, content fixes and upgrades.

Built for Teams That Need to Trust the Answers

Emly is designed around control: where it runs, what it answers from, and which model it uses.

Runs in Your Cloud

Content, search index, conversations and audit log stay in storage you control on AWS, Azure, Google Cloud or your own servers.

Citations on Every Answer

Each answer links to the document it came from, so customers and staff can check it.

Your Model, Your Key

Use any OpenAI-compatible model, including Azure OpenAI or one you host yourself, under your own agreement.

Open Source, No Lock-In

Your team can read every line, run its own scans and keep running it if you ever part ways with us.

Conversational AI We've Already Delivered

Datoin has been building production AI for more than 15 years, including assistants that take routine questions off support teams.

GenAI Support Automation

Automating a Service Desk with a GenAI Assistant

A multilingual assistant took routine calls off the front desk and improved appointment handling.

70%Response time reduction
60%Call volume reduction
$82KCost savings

Before You Choose

Emly is an open-source AI assistant platform. Datoin deploys it inside your own cloud, connects it to your content and the language model you choose, and supports it in production. Every answer cites the document it came from.

Launch in 14 is for customer-facing support: an assistant on your website or in your product that answers from your help center and docs. Private AI Workspace is for internal teams such as IT, HR and Compliance, with company sign-in, separate assistants per department and a security pack for your reviewers.

In your own cloud account (AWS, Azure or Google Cloud) or your own servers. You can use any language model with an OpenAI-compatible API, including Azure OpenAI and models you host yourself, with your own API key.

Each package is a fixed-price setup that we quote after a 30-minute scoping call. There's no per-seat licence, and you pay your model provider directly. An optional monthly Care Plan covers content reviews, fixes and upgrades.

See It Working on Your Own Content

Send us your website URL and we'll build a working assistant on your public pages within 48 hours, free.