What we do

Chatbots & AI assistants

A chatbot is only as good as what it is allowed to say and what happens when it runs out of answers. We build support and sales assistants for customers and for your own team: grounded in your knowledge base, citing the source behind each answer, and working to agreed rules for when a conversation goes to a person and which person it goes to. They answer around the clock, and hand over into a shared inbox where the team can see the thread. The nearest example in our work is an automotive marketplace's staff console: Instagram, Messenger, and WhatsApp in one shared inbox, with a trilingual AI assistant alongside it.

Book a call Send the details →

ChatwootWhatsAppInstagramMessengerNext.jsSupabase

Scope

What's included, and what isn't.

Included

Support assistants that answer customer questions, so your team is not answering the same ones by hand.
Sales assistants that answer the questions a buyer has before they enquire, and pass the conversation to your team once there is a real lead.
Knowledge-base grounding with citations: answers come from your material, not from whatever a model happens to know, and each one names its source so a person can check it.
Human handoff and escalation rules: which conversations go to a person, which person, and with the thread attached so the customer does not start again.
A shared inbox for Instagram, Messenger, and WhatsApp, so a conversation the assistant hands over lands where the team already answers, rather than across three apps.
Assistants for your own team, inside the console or inbox they already work from, answering from internal material.
Assistants that answer in more than one language, where your customers write in more than one.

Not this

Outbound nurture and follow-up sequences: those sit under GoHighLevel & automation. This page is about answering the conversations people start.
AI features inside a product, document extraction and classification, and custom model work: those sit under AI development.
An assistant that replaces your support team. It answers what it can from your material and hands the rest to a person.

Built and running

Products we've shipped that do this.

Automotive marketplace

A car platform and the console that answers every customer

An automotive marketplace for buying, selling, sourcing and importing cars, with a 121-brand catalogue and four lead forms. Behind it we built the staff console: one shared inbox for every DM, a lead queue on a reply clock, inventory, a viewing diary and an AI assistant that works in three languages.

None →100%Web enquiries recordedEvery enquiry from the website is now recorded with a reference; before, none were.

100%DMs in one inboxAll Instagram, Messenger and WhatsApp messages are answered from one shared inbox.

4 → 1Lead forms into one queueFour lead forms (sell, source, import and enquire) feed a single queue with a reply clock.

Shared inboxAI assistantLead CRM

Next.js
React
Supabase
Cloudflare Workers
Chatwoot

Straight answers

Questions about Chatbots & AI assistants.

What stops it making things up? +

Grounding and rules, not a promise. It answers from your knowledge base and cites the source for each answer, so anyone can check where an answer came from, and the escalation rules send questions your material does not cover to a person.

When does a person take over? +

When your rules say so, and we agree those rules with you: the questions it should never answer, the moment someone asks for a person, the conversations worth a salesperson's time. The conversation then lands in the shared inbox with its history, with whoever the rules name.

Which messaging channels do you work with? +

The shared inbox we have shipped brings Instagram, Messenger, and WhatsApp into one place, and that is where a conversation lands when it goes to a person. If your customers write to you somewhere else, tell us on the call and we will say plainly whether it fits.

How is this different from the AI nurture in your GoHighLevel work? +

Nurture is outbound: the CRM writes and sends a follow-up to a lead on a schedule, then waits for a reply. A chatbot is inbound: someone asks a question, the assistant answers it, and hands over when it should. Outbound follow-up in GoHighLevel sits under GoHighLevel & automation.

Do we actually need a chatbot? +

Not always. If the questions are few and the answers rarely change, a good FAQ page does the job for less, and we will say so on the call. And if what you need is AI inside the product rather than an assistant people talk to, that is AI development.

Start a project

Let's build something you can run your business on.

An automation, a new app, a website, an AI feature or chatbot, or a rescue: start with a call. Tell us the problem; we'll tell you the smartest way to solve it.

What happens on the call Thirty minutes. You describe the problem and the stack. We tell you what we'd build, roughly what it costs, and whether we're the right fit. If we're not, we'll say so.

Book a call See our work →