Case Studies | FinTech

Bringing financial data to life

AI Agent and Financial Data Automation

About

A FinTech business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The challenge had three focus areas.

Better, safer function calling

Function calling on a local Mistral model needed to become more accurate and lower latency, with guardrails for safe and compliant tool execution, by migrating to Amazon Bedrock and choosing the right model and prompting strategy.

Connecting the tools customers use

The client's users live in QuickBooks and Xero. The platform needed reliable connectors to both so it could read and write accounting data through a single, unified schema.

Understanding bank statements automatically

Raw bank transactions had to be classified and categorised with context awareness, distinguishing, for example, a grocery purchase from a fuel purchase at the same merchant, with confidence scoring and a human-in-the-loop review for the uncertain cases.

Solution

Cloud Combinator delivered the engagement as three parallel workstreams, sequencing the AI techniques from lightest to heaviest so cost and complexity were only added when justified.

75%+

Classification accuracy target at initial deployment (target)

2

Accounting platforms integrated: QuickBooks and Xero

3

Workstreams delivered in one engagement

By the numbers:

  • 75%+ - Classification accuracy target at initial deployment (target)
  • 2 - Accounting platforms integrated: QuickBooks and Xero
  • 3 - Workstreams delivered in one engagement
Changes

The engagement moved the client's function calling onto Amazon Bedrock with guardrails, delivered QuickBooks and Xero integration through an unified schema, and added a bank-transaction classification workflow with confidence scoring and human review, all integrated with the client's existing pipeline.

  • Migrated to Bedrock — Function calling moved from a local Mistral model to Amazon Bedrock with guardrails for safe, compliant tool execution.
  • Smarter tool selection — An updated tool registry and tuned prompting improve the relevance of the agent's tool choices and reduce latency in selection.
  • Connected accounting — QuickBooks and Xero are integrated through an unified data schema with functions for invoices, expenses, accounts and transaction sync.
  • Context-aware categorisation — Bank transactions are classified with a merchant-level taxonomy, confidence scoring and a human-in-the-loop fallback for low-confidence cases.
  • Handover — The client received the migrated function-calling architecture, connector integrations and documentation, running within its existing pipeline.

With its agents on Amazon Bedrock and its accounting and bank-statement data understood automatically, the client has a foundation it can extend with retrieval-augmented generation or fine-tuning as accuracy needs grow.

AWS Stack

Amazon Bedrock

Model selection and prompt optimisation, with retrieval-augmented generation available as a next step, for improved tool selection.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

We use cookies

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.