AI support that resolves, then hands over.
Modulus Labs AI builds chatbots and support agents that answer from your own content, act in your order, booking and help desk systems, and pass the conversation to your team with a summary when a person should take it.
- Web chat · WhatsApp · Email · Your help desk
- Your help content · Policies · Live order and booking data
- Resolution · Handover quality · Accuracy · Customer rating
- NVIDIA Inception member since July 2026
- Within one business day
Support that finishes the job.
Most companies need one or two of these at first. We start where your queue is longest and the risk is lowest.
-
Self-service agents
Answer status, how-to and policy questions, and make simple changes in your systems within the limits you set. On WhatsApp too: WhatsApp AI agents.
-
Agent assist
Draft replies, summarise the case and find the right policy for your team, with a person reading every reply before the customer does. The lowest-risk place to start.
-
Routing and triage
Classify each message, set its priority and send it to the right queue, with the details a person needs already collected.
-
Conversation review
Read conversations to find the top reasons for contact, failed answers and gaps in your help content, so the fixes go where they count.
Grounded, limited, measured.
What separates support customers trust from a chatbot they try to escape.
-
Answers from your sources
Replies come from your help content, policies and live systems, never the model's general knowledge, and the agent says so when it can't find an answer. Retrieval that says "I don't know".
-
Handover written as code
The rules for passing a conversation to a person run before every reply, so no customer can talk the agent out of them. Our AI customer service guide.
-
Tested on real tickets
Real past conversations from your queue, including the awkward ones, scored before every change and reviewed weekly after launch. Evals before features.
Systems already running.
Our WhatsApp agent handles 70% of a solar company's monthly sales and hands buyers to the sales team when it should, with a 4.8 out of 5 customer rating and 99% uptime. The same core of knowledge, system access, rules and tests can sit behind web chat and email.
-
Do we need a chatbot or an AI agent?
A chatbot answers from your content. A support agent also acts in your systems: it looks up the order, reschedules the booking or updates the address. Most support teams start with answers and add actions one at a time, each with its own limit.
-
When does it hand over to a person?
When the customer asks for a person, after two failed attempts at the same question, when nothing relevant is found, when an action is above a limit, on complaints or signs of distress, and when one of your systems is down. These rules run in code before the reply, and the person receives a summary so the customer never repeats themselves.
-
How do you measure whether it works?
On resolution, not containment: a conversation counts only if no person was needed and the customer didn't come back about the same issue within seven days. We also track escalation reasons, handover quality, customer satisfaction against your human baseline and accuracy on a weekly reviewed sample.
-
How long does a project take?
A focused system usually reaches production in six to ten weeks; larger integrations take three to five months. Discovery comes first, so you see the plan before any build.
-
Who owns what we build?
You do: the code, prompts, evaluation sets, infrastructure and documentation. When we hand over, your team can run and extend it.
Hiring a team to build one? Read the questions to ask an AI development company and how to test an AI system before you accept it.