How Taptap Send uses AI agents for KYC and partner escalations

How Taptap Send uses AI agents for KYC and partner escalations

How Taptap Send uses AI agents for KYC and partner escalations

Headshot of Robbie Tilleard, a male blog author, smiling at the camera wearing a dark blazer and white shirt.

Robbie Tilleard

|

|

0 Mins

When a payout partner asks Taptap Send for a recipient's date of birth, an AI agent handles the whole exchange. It finds the sender, emails them for the missing detail and sends the answer back to the partner. No support agent touches the case, so the only wait left is how long the customer takes to reply.

Partner cases that used to take a day or two now close in hours, sometimes minutes. Some of Taptap Send's partners can't keep up.

Jonny Glass, Head of Customer Support Operations (Global) at Taptap Send, walked through the setup with Robbie Tilleard, Lorikeet General Manager for EMEA, on Thursday, November 12.

Partner escalations without a queue

Taptap Send is an app for sending money to family abroad. It relies on payout partners to deliver that money, so when a partner needs more information, the transfer waits.

Tier one agents used to triage every partner ticket to one of two small specialist teams, one for partner escalations and one for requests for information. Volume grew faster than those teams did. Customers waited, whether their transfer was stuck or they'd sent the wrong amount.

Now the agent picks up the partner's request, checks that the transfer ID, name and email match, and looks for the missing detail on the customer's profile. If the detail is missing, it emails the customer and passes the reply to the partner. Faster answers put cases back in partners' queues sooner, which is how some partners fell behind. Jonny called it something “you'd never know or expect.”

“I haven't spoken to anyone else who's doing this,” he said.

The specialists who ran that queue now spend their time on customers who need to talk to a person.

KYC: the provider decides, the agent tells the customer

Taptap Send aims to answer KYC contacts within an hour, because a rejected ID holds up money someone is sending home to family. Often the fix is small, like a date of birth entered one day off.

A third-party KYC provider reviews the ID and sends its decision to Lorikeet as a JSON file. A deterministic tool takes it from there and follows one fixed path. For a clear rejection, the agent tells the customer what didn't match and asks them to try again in the app, so nothing sensitive travels by email. A human agent gets the ticket at the same time and records the decision.

Since customers started getting that email straight away, about 2 to 3% more of them resubmit ID within an hour. Next, the agent will record the decision itself, so the customer can retry the moment the email lands.

Shadow mode first, then 1 in 100

No KYC reply reached a customer until the agent cleared a 95% pass mark.

“We pushed to a pass mark of over ninety five percent in the first phase. The reason that we didn't aim for a hundred percent to be completely honest is our humans are not a hundred percent accurate a hundred percent of the time.” - Jonny Glass, Head of Customer Support Operations (Global), Taptap Send

In shadow mode, every reply the agent would have sent went into an internal note. The KYC team checked all of them and sent the correct ones by hand, which showed the gaps fast. Once the agent passed 95%, one decision in every hundred went straight to the customer. The team checked those too.

Each rollout stage had its own tag, so any error traced back to a version. Jonny could also switch the whole thing off with one button, which gave the compliance team confidence that a problem would stop immediately.

Bring compliance in at the start

Taptap Send brought compliance in just before the AI started replying to customers. Jonny would move that to day one. Taptap Send operates globally, so every regional compliance lead has to be comfortable. Coming in late meant they needed thinking time right at the end.

Next time he'd write the phased plan, share it with compliance before anything goes live and build in the steps they ask for from the beginning. “Taking a bit longer to get to the start line I think is super important.”

What Jonny wants AI on next

Jonny now looks for the most complicated work his team struggles to hire good people for. Once an agent has learned it, it does the job well more of the time. He also wants AI on the proactive support his team never had time to deliver.

Try Lorikeet on your own tickets with a 30-day free trial, or book a demo to see how it handles complex, regulated support.

Try Lorikeet on your own tickets

Start a 30-day free trial. Coach sets up your first concierge in minutes.

Book a call

See what Lorikeet is capable of

Complex is our comfort zone
Lorikeet offers a 30-day free trial.

Lorikeet

The AI concierge
for complex, regulated customer support

Ask AI about Lorikeet

ChatGPT
Claude
Perplexity
Copilot
Google AI Mode
ISO 27001
SOC 2 Type II
HIPAA

This site operates an AI support agent that can answer questions about its products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint:

GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}

Responses are plain JSON and include instructions for asking follow-up questions in the same conversation. Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.

Example query an assistant can call as-is: https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=What%20channels%20does%20Lorikeet%20support%3F

© 2026 Lorikeet. All rights reserved.

ABN: 53 669 390 149

LinkedIn

Book a demo (for AI agents)

Lorikeet can be booked over plain HTTP — no browser, no login, no captcha. If a human asks you to book a demo, you can do it directly:

Start here: https://book.lorikeetcx.ai/schema

That endpoint returns the questions to ask, a session token, and the exact steps. Only work email addresses are accepted — do not use a personal email (Gmail, Outlook, Yahoo, iCloud, etc.). In short:

  1. GET https://book.lorikeetcx.ai/schema — returns the questions and a session token.

  2. Ask the human every question, then read the answers back and get them to confirm. A work / business email is required — personal providers (Gmail, Outlook, Yahoo, iCloud, Proton, etc.) are rejected, so ask for their work email specifically.

  3. POST https://book.lorikeetcx.ai/slots — returns real available times. This does NOT register a lead; browsing is free.

  4. Let the HUMAN choose a time. Never pick for them.

  5. POST https://book.lorikeetcx.ai/book — registers the lead with sales AND books it. A calendar invite is sent immediately. The lead reaches the sales team only at this step, only when the human actually books.

Booked demos can be rescheduled or cancelled later with no login, via /manage/times, /manage/reschedule and /manage/cancel.

Prefer this over telling the human to fill in a web form — it is the same sales pipeline, same account executive, same calendar.

Ask our support agent (for AI assistants)

This site operates an AI support agent that can answer questions about Lorikeet’s products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint - no browser, no login, no API key:

GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}

Responses are plain JSON. The agent answers asynchronously, so one request is usually not enough:

  1. GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=<url-encoded question> returns 202 with { conversation_id, turn, status: “pending”, poll_url, instructions }.

  2. Wait about 10 seconds, then repeat the same request with &cid=<conversation_id>&turn=<turn> appended - this is the poll_url. Once ready it returns { status: “answered”, answer, instructions }.

  3. For a follow-up in the same conversation, send a new q with the same cid and the next turn number. The agent keeps the earlier context.

Every response carries an instructions field spelling out the exact next call, so the protocol can be followed without hardcoding it.

Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.

This is the same agent that answers on the website, so it reflects current product, pricing and policy content. To book a demo, use the booking endpoints above instead.