Case Studies | FinTech

A RAG-powered assistant for ad campaigns

AI Ad-Campaign Assistant

About

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

Challenge

The engagement concentrated on three focus areas, each essential to a trustworthy conversational assistant.

Guiding merchants through campaign setup

Shopify merchants needed conversational, step-by-step help to create and run optimised Google Ads and Digital Audio campaigns. The assistant had to understand what a merchant was asking and lead them through the process rather than simply answering in isolation.

Grounding answers in real knowledge

Generic model output was not enough. Responses needed to draw on the client's own knowledge base so that guidance was accurate and specific, which pointed to retrieval-augmented generation over a vector database rather than an unaided language model.

Trust, safety and compliance

Because the assistant speaks directly to customers, it needed guardrails that define acceptable use and boundaries for the model, and a design that follows the client's compliance and regulatory requirements.

Solution

A merchant's question enters through a chat interface and Amazon Lex, which captures intent and hands off to an AWS Lambda function acting as the chat orchestrator. The orchestrator queries an Amazon OpenSearch vector database to retrieve the most relevant passages from the client's knowledge base, then combines that context with the user's query into a single prompt.

5+

AWS services orchestrated into one conversational pipeline

RAG

Retrieval-augmented generation grounding every answer in the knowledge base

$36.6k

Projected annual AWS run cost at modelled usage (projected)

By the numbers:

  • 5+ - AWS services orchestrated into one conversational pipeline
  • RAG - Retrieval-augmented generation grounding every answer in the knowledge base
  • $36.6k - Projected annual AWS run cost at modelled usage (projected)
Changes

The proof of concept met its scope: it demonstrated that Amazon Bedrock, paired with retrieval-augmented generation over an Amazon OpenSearch vector database, is a suitable backend for the client's AI webchat, giving merchants conversational, knowledge-grounded guidance through campaign setup.

  • Conversational campaign guidance — An Amazon Lex-fronted chatbot understands merchant intent and walks Shopify sellers through setting up and running Google Ads and Digital Audio campaigns.
  • Grounded answers — Retrieval-augmented generation over an Amazon OpenSearch vector database pulls relevant knowledge-base content into each prompt, so responses are accurate and specific rather than generic.
  • Conversational memory — Amazon DynamoDB retains context across the conversation, letting the assistant follow a multi-step campaign setup without losing the thread.
  • Guardrails and compliance — Acceptable-use boundaries were defined for the model and the build followed the client's compliance and regulatory requirements.
  • Handover — The solution was deployed into the client's AWS environment with an UAT phase and knowledge-transfer sessions, with acceptance tied to proving Bedrock a suitable platform for continued development.

With Amazon Bedrock and a retrieval-augmented architecture validated, the client is positioned to continue developing its AI marketing assistant on AWS and extend grounded, conversational guidance to more Shopify merchants.

AWS Stack

Amazon Bedrock

For large language model generation of campaign guidance grounded in retrieved knowledge.

Amazon Lex

For the conversational front end that captures merchant intent and utterances.

Amazon OpenSearch Service

For the vector database that powers retrieval-augmented generation.

Amazon DynamoDB

For low-latency conversational memory across the session.

AWS Lambda

For serverless orchestration of the chat pipeline between services.

Amazon S3

For secure storage of the knowledge-base data source.

AWS IAM

For access control and permissions across the pipeline.

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.