×

Lorikeet Overview for TerraPay

A team of AI agents that resolves partner status queries end-to-end, across 500 partners, every corridor, every channel.

Prepared for Ashok Gajarajan, Rupee, Kameshwar & Ashirbad · 15 May 2026
Robbie Tilleard, Lorikeet

The Opportunity

Today a 100-person ops team sits in the middle, manually pulling fragmented data to answer 30k partner queries a month. Imagine the alternative.

Integrated data

One layer that reads across the transaction DB, corridor rules, payout-partner status, and compliance state. No more bouncing between five ops screens to answer one query.

Partner problems automated

Status diagnosis, failed-transaction reroutes, and proof-of-delivery chases resolved end-to-end by the agent. Replies land in the channel the partner used, every time.

Proactive outage management

Agent detects payout-partner outages, flags every stuck transaction, and notifies each affected partner in their channel of choice. Hundreds of inbound queries become a handful of outbound notifications.

Team on the work that matters

The 100-person ops team stops being a manual relay between partner inboxes and TerraPay's systems. They move to partner success, network growth, and the exceptions only humans can resolve.

State of Play at TerraPay

100-person ops team handling ~30,000 partner status queries a month across ~500 partners globally.
Manual work, every query — ops looks up the transaction ID, queries multiple internal systems, reconciles compliance / payout / tech state, translates binary status codes, then replies.
Sitting in the middle — partners on Slack, WhatsApp, email, Teams, social on one side; fragmented internal data on the other. Today the team is the integration layer between them.
Diagnosis isn't one-shot — "where is my money" can be a compliance hold, a payout-partner outage, a tech failure, a network disconnect, or insufficient funds. The team has to triage before they can answer.
Corridor rules differ by country — Pakistan is real-time, Bangladesh has bank cutoffs and Fri/Sat holidays, India runs across IMPS, NEFT, and RTGS. The right answer depends on which corridor the transaction is on.
Beyond status: operational work — failed IMPS gets re-pushed via NEFT, proof of delivery has to be chased from payout partners, wrong-beneficiary cases need reversal. All by hand today.

Three scenarios in the demo

Synthetic transaction data, real workflows, real AI agent. We'll run all three live on the next slide.

01

Delayed real-time transfer

Taptap Send asks where TPV-4471 (£100 to Vietnam) is. The AI agent looks up the transaction, reads the GB-VN corridor rules, checks the payout partner's network status, identifies a partner-side stall, and quotes the expected window.

Status query — the 30k-a-month volume driver.

02

Failed transaction reroute

TPV-5582 failed on IMPS to India. The AI agent confirms the failure is rail-level (not compliance), checks the policy, proposes NEFT, and executes the reroute after the partner confirms.

Operational write action with a policy gate.

03

Non-credit complaint, proof of delivery

TPV-6603 marked settled to Kenya, customer says no funds. The AI agent checks for an automatic proof of delivery, finds the corridor doesn't expose one via API, opens a request with the payout partner, and quotes the SLA back to Taptap Send.

Multi-step orchestration across two partners.

The same sandbox also handles cutoff-corridor queries (Bangladesh banking holidays), pure delay diagnosis (compliance vs payout vs tech vs insufficient funds), and wrong-beneficiary reversals — happy to demo any of these on request.

Try a ticket against the live AI agent.

Mock partner-portal UI on the right, with the real Lorikeet chat widget embedded against a TerraPay sandbox. Synthetic transactions, real AI agent, real tools. The router defaults to Taptap Send Ops in this sandbox — just paste a query.

Try: Vietnam status query

One of our senders is asking about transaction TPV-4471 to Vietnam — £100 sent yesterday afternoon, not yet received. Where is it?

Try: IMPS reroute

Transaction TPV-5582 to India failed via IMPS. Can you push it through on NEFT?

Try: non-credit complaint

Transaction TPV-6603 to Kenya was marked successful three days ago, but the customer says the funds never arrived. Can you confirm proof of delivery?

Answers cite synthetic transaction state plus your real corridor rules and proof-of-delivery process. Wrong-beneficiary reversals and cutoff-corridor queries run on the same sandbox — happy to demo on request.

Team of AI agents in action

Two patterns where one AI agent isn't enough.

Inbound — "Where is my money?"

1

Partner

Emails or messages asking about a delayed transfer.

↓

2

Concierge AI agent

Picks up the query and identifies that payout-partner data is needed.

↓

3

Specialist AI agent

Contacts the payout partner via API or message.

↓

4

Specialist AI agent

Validates the response and hands it back to the concierge.

↓

5

Concierge AI agent

Replies to the partner in the channel the query came in on.

Outbound — Outage management

1

Signal

Outage detected on a payout partner via tool feed or webhook.

↓

2

Triage AI agent

Finds every stuck transaction and the affected partners behind them.

↓

3

Notification AI agent

Pushes a status update to each partner — Slack, WhatsApp, email, Teams.

↓

4

Partner

Replies in their own channel asking for follow-ups.

↓

5

Concierge AI agent

Handles each follow-up in-channel with the current outage state.

Any task with these three properties can run on AI agents

Lorikeet integrates data and communicates across multiple channels. That's the whole pattern.

01

Grabbing data

Read across systems — transaction state, corridor rules, compliance posture, partner network status.

02

Confirming with partners

Reach external partners, validate their response, retry if needed, hand the answer back.

03

Communicating with customers

Reply in the partner's channel of choice — Slack, WhatsApp, email, Teams, voice.

Lorikeet runs the orchestration. Your team owns the exceptions.

Appendix

Lorikeet overview, the team, investors, customer voice, architecture, POC plan, security, and pricing.

Lorikeet is building the leading AI concierge for complex businesses

Lorikeet exists to give every company the ability to deliver a universal concierge to their customers. Customers should demand better support, Lorikeet's platform solves issues by understanding customer needs and taking actions for them, 24/7, via the channel of their choice.

Re: Status of TPV-4471, Vietnam transfer for Anjali S.

Lorikeet

To: Taptap Send Ops

Hi team, TPV-4471 is currently held at our Vietnam payout partner — their domestic switch is reporting intermittent timeouts since 08:14 UTC. Funds are safe.

Per the VN real-time corridor SLA, expect delivery once the switch clears (typically <2h). I'll send a follow-up the moment we see the credit confirmation.

Email

Lorikeet

0:08

"Hi, this is TerraPay's partner concierge. I'm calling about transaction TPV-5582 — the IMPS attempt failed at 09:42 IST. Want me to re-push via NEFT? Delivery in two to three hours."

Voice

Status of TPV-6603 to Kenya? Customer says they didn't receive it.

T

Marked delivered 3 days ago, but the Kenya corridor doesn't expose automatic proof of delivery via API. I've opened a proof-of-delivery request with the payout partner now.

How long for confirmation?

T

Partner SLA is 24h. I'll send the proof of delivery straight back to you here once it lands.

Chat

Our people have decades of experience with AI and building great products

CEO and Co-founder

Steve led product teams at Stripe and Watershed, building tooling to enable complex processes like carbon accounting and financial reporting at scale.

CTO and Co-founder

Jamie was a research tech lead at Google Brain, leading research on factual grounding in large language models. Third named author on Google's breakthrough 2022 LaMDA paper, and fourth named author on the 2020 predecessor Meena paper.

Our team comes from leading companies applying AI and driving large-scale corporate transformation

Lorikeet is backed by top global investors

We've raised over $50M from investors who backed Nubank, Klarna, Canva, Airwallex, Supabase, Midas, Clearscore and more.

Announcing our

$35M

Series A

Led by QED Investors

Blackbird

Square Peg

Capital49

Skip Capital

Airtree

Operator Partners

www.lorikeetcx.ai

Lorikeet consistently outperforms other vendors in the market

"We tested AI solutions head-to-head and Lorikeet was a winner in every metric."

Lindsay Boland

Product Manager, Flex

"We ran POCs with Fin AI, Decagon... Lorikeet was a clear winner."

Jiaona Zhang

Former CPO, Linktree

"Considered Sierra and other well-known players... [Lorikeet] could handle nuance, de-escalate emotional moments, and follow SOPs without breaking a sweat."

Jessica Mishlove

Head of Customer Relations, Arbor

Read case studies

Already automating partner support for cross-border payments leaders

Journey of a Support Ticket

How Lorikeet's AI agent resolves customer issues end-to-end

1

Customer sends message

Via chat widget, email, WhatsApp, voice, or ticketing system

2

Channel Intake

Mode (chat / email / voice / SMS) · Escalation rules · Auto-close config

3

Brand + Config Loaded

TerraPay voice · Partner-facing tone · Per-corridor rule set

4

Inbound Guardrail

Evaluate message → pass to triage, or escalate immediately

5

Intent Classification

AI matches customer intent against all available workflows

Matched Workflow

Run response SOP

FAQ Fallback

Search knowledge base

No Match

Escalate to human

6

Workflow Executes

Structured decision graph or Natural Language agent follows step-by-step SOP

Tools

Txn DB · Corridor rules · Partner status

Knowledge

Proof of delivery · Reversal policy · SOPs

Actions

Reroute · Proof-of-delivery chase · Tags · Reply

7

Outbound Guardrail

Check AI response against policies → pass or escalate

8

Reply Sent to Customer

Brand voice · Formatting · Citations · Side effects (tags, CSAT, Slack)

If customer replies → loops back to step 5 with full context

If no reply → auto-resolved after idle period

Proof of concept

2–4 weeks, offline, free, to prove Lorikeet handles real TerraPay partner status queries end-to-end across the 2–3 highest-volume corridors
We'll mock the partner data to simulate production situations, rather than integrate with your platform on day one
Zero risk, if it doesn't meet the bar Rupee's team sets, no obligation to proceed
Live pilot with one partner, e.g. Taptap Send for the UK–Asia corridors, then scale by partner cohort once we're hitting the resolution rate you want

Train and test AI

Integration scoping

Legal / privacy / security review

Week 0–1

We will:

Train Lorikeet's tone and brand from your guidelines
Run response testing over 20–50 example tickets
Plan training based on your SOPs

We need:

Tone of voice, SOP, ticket corpus
Fast feedback on tickets

Week 1–2

We will:

Build, test and iterate on 1–2 agreed use cases
Share learning materials with your team
Produce documentation to aid implementation

We need:

Continued feedback on AI response quality

Week 2–3

We will:

Run Lorikeet 101 training session with your champion
Complete any necessary security and legal reviews
Agree on terms of contract including cost

We need:

Tech lead to align on integration plan

Week 3–4

We will:

Complete security and legal reviews
Kick off integration sync

We need:

Agree on commercial terms and sign

Lorikeet is built securely for enterprise

Compliance-grade architecture, day one. Designed for FCA-regulated fintechs and cross-border payments infrastructure handling sender PII and transaction data.

SOC 2 Type II
& ISO 27001

Both certifications independently audited. Your data is always protected and is never used for training.

GCP + VPC

Built on Google Cloud, running in a virtual private cloud. Audit-grade logging end-to-end.

Ex-Stripe security

Founding engineering team includes an ex-Stripe Staff Security Engineer. Security reviewed by industry leaders.

Handles sensitive PII at scale

Already deployed with FCA-regulated cross-border payments players (Airwallex, Taptap Send) handling sender PII, transaction data, and partner-level commercial info at scale.

For TerraPay's data posture: sender PII, transaction IDs, partner-level commercial data, and compliance hold internals all need strict handling and audit trails. We're happy to walk your security and compliance leads through our controls, data residency by region (US / EU / AU), access controls, and incident response.

View trust center →

You only pay for resolutions

Aligned incentives by design, we both win when tickets are resolved.

Pay only for resolved tickets

Lorikeet charges per ticket successfully resolved, as determined by you. If we don't deliver, you don't pay.

No platform or per-seat fees

No setup costs, no per-seat licenses, no monthly minimums. The only line item is resolutions.

Volume discounts

Per-resolution price decreases as volume grows. Available across email, chat, voice, and SMS.

Aligned incentive structure

You buy credits from Lorikeet up front
Credits are only consumed when the AI reaches the desired outcome
You don't pay if the AI fails and the case is passed to a human
Credits sized to your forecast volume, buy top-up packs at a discount mid-contract, no surprise overages

At TerraPay's volume: ~30,000 status queries / month puts you firmly in our enterprise tier. Exact per-resolution price depends on the workflows we scope during the pilot, with volume-based step-downs as more corridors come online. We'll model the breakeven against the 100-person ops team during the next session.

Robbie Tilleard

robbie@lorikeetcx.ai