Westpac runs five AI agents in its home-loan and credit-card lending
In plain words
Westpac plugged five AWS agents into the core of how it processes loans and cards.
What happened: More than 150,000 hours saved.
Source: itnews.com.au ↗
How to do it yourself
Our suggested steps. The company didn't publish theirs.
- 1
Write your application checklist: documents required, income rules, red flags.
- 2
Collect 10 past applications (anonymised) to test on.
- 3
Run each through Claude with the prompt below and compare with what your team decided.
- 4
Adjust the checklist where the agent disagreed for good reasons.
- 5
Use Make/n8n to send new applications' documents to the agent and post its summary to your team channel.
The instructions to give the AI
Paste this into ChatGPT (the free plan is fine), swap the [brackets] for your details, and test it on a few real examples.
This is one part of the setup. It won't connect to your phone, inbox or other apps on its own; the steps above cover that.
Agent instructions (prompt)
You pre-check lending/credit applications for [business]. Checklist: [paste] For the application below: 1) list which documents are present/missing, 2) extract income, expenses and existing debts into a table, 3) apply each rule and mark pass/fail, 4) list red flags, 5) write a 5-line summary for the credit officer. Do not make the final decision. Application: [paste]
[Highlighted] = replace with your own details before you run it.
What you should get back: "Present: payslips, ID. Missing: 3 months bank statements. Flag: income on form ≠ payslip ($82k vs $76k)."
Written by Quickly and Good, inspired by this case.
More detail: tools, prices and what you'll need
The company used: Amazon Bedrock AgentCore
You could use: Claude (or ChatGPT) + your loan/credit application checklist; Make or n8n to route documents
- Amazon Bedrock
AWS service for building AI apps and agents on top models.
See pricing · Official site ↗
- ChatGPT
General AI assistant: writes, summarises and answers from files you give it.
Free plan; paid plans available · Official site ↗
- Claude
AI assistant for writing, analysis and long documents; can connect to your tools.
Free plan; Pro $20/mo (monthly) · Official site ↗
- n8n
Workflow automation that connects apps and adds AI steps; can be self-hosted.
Free to self-host (Community Edition); cloud: see pricing · Official site ↗
Prices from each tool's own pricing page, checked Oct 2026.
What you'll need
- ☐An automation tool (n8n, Make or Zapier)
- ☐Connections to each app involved (email, CRM, sheets, calendar)
- ☐An AI model account/API key
- ☐The agent instructions (prompt)
- ☐Error handling: what happens when a step fails
- ☐Testing with real data for a week
- ☐A human review step before anything goes out to customers
Can I do this myself?
Yes if you have built a Zapier/Make/n8n workflow before. If those words are new, hire a builder.
Similar cases
CMA CGM's AI voice agents resolve 46% of cargo release calls
More than 46% of cargo release enquiries are resolved by the AI agents, and about a third of 4,000 customers called by the sales agent moved on to active discussions.
ElevenLabs' own AI support agent resolves 65% of its tickets
65% of tickets resolved end to end, first reply in about 60 seconds instead of hours, and 70% positive ratings.
Telstra lets an AI agent sort refunds while customers wait for repairs
Refunds are worked out and paid automatically once service is back, which takes admin off 1,000+ staff.