We build AI and human customer support that businesses can trust and control
Support software should answer accurately, hand over to a person cleanly, and stay recognisably yours. That is the whole idea behind the platform.
Why this exists
The gap between AI that overpromises and human support that cannot scale
Most teams end up choosing between two imperfect options. AI-only tools answer instantly but will happily invent an answer they were never given, which costs more trust than they save in time. Human-only support is accurate but expensive to staff, and the same handful of questions arrive again and again at every hour. We started here because neither side of that trade-off is necessary: an AI agent restricted to a business's own approved content can carry the repetitive volume, and a person can take the thread the moment judgement is needed.
- AI answers drawn from your own content, not from a general model's guesswork
- A visible escalation path instead of a confident wrong answer
- One thread, so the person taking over sees everything the visitor already said
What we believe
Three principles that shape what we build
These are the decisions we keep returning to when a feature could go either way. They are also the fastest way to tell whether the product is doing its job.
- AI should be grounded in real content, not guesswork — if the answer is not in your material, the right move is to escalate.
- Human support should never be a paywall — unlimited human live chat is part of every plan, including the free one.
- Businesses should own their brand, not ours — white-label control means the experience looks like your product.
We are an early-stage product and we would rather describe what the platform does today than what it might do later.
See how it behaves with your own content
Start free with unlimited human live chat, then switch the AI on when your knowledge base is ready.