Case Studies | SaaS B2B

Proving Amazon Bedrock for agentic AI

AI ML Assessment for Agentic Workloads

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The challenge had three focus areas, all aimed at de-risking a move to AWS before large-scale migration began.

Confirming Bedrock for agentic and generative AI

The central question was whether Amazon Bedrock could host the client's agentic and generative AI services with comparable behaviour. This called for testing the client's existing prompts against Bedrock to check that similar results were obtained before any dependency was committed.

Mapping the wider Azure environment

Beyond the AI layer, the assessment had to work through access controls, function apps, and the split frontend and microservices, translating Azure IAM to AWS IAM and identifying equivalent AWS services for each component so nothing was missed in a later migration.

Producing a credible, costed migration plan

The engagement needed to end in a migration plan that reflected the client's business and technical requirements, backed by an indicative AWS cost model so leadership could plan the move with eyes open.

Solution

Agentic AI means the application does not just answer a single prompt; its AI agents carry out multi-step tasks, calling services and tools to complete a workflow. In this assessment, Amazon Bedrock provides the inference plane that powers that content generation, so the client can use foundation models through a managed AWS service rather than running model infrastructure itself.

~$15.5k/mo

Indicative AWS run-rate for the assessed target architecture

2 Lambdas

Isolating the backend microservices from the agentic AI flow

Prompt

Existing prompts tested against Amazon Bedrock for comparable results

By the numbers:

  • ~$15.5k/mo - Indicative AWS run-rate for the assessed target architecture
  • 2 Lambdas - Isolating the backend microservices from the agentic AI flow
  • Prompt - Existing prompts tested against Amazon Bedrock for comparable results
Changes

The assessment gave the client the evidence and the plan it needed to move forward with confidence: a tested view of Bedrock's fit for its agentic AI, a mapped and costed target architecture, and a migration plan for the entire environment, all delivered as a working reference deployment ready to hand over.

  • A tested Bedrock assessment — The client's agentic and generative AI workloads were assessed against Amazon Bedrock, including prompt testing to check for comparable results.
  • A mapped target environment — Access controls, function apps, and the split frontend and microservices were mapped to AWS services, with Azure IAM translated to correctly scoped AWS IAM.
  • A working reference deployment — A stack of two Lambdas from Amazon ECR, API Gateway, RDS for PostgreSQL, AWS Amplify, Amazon SQS, and Amazon Managed Grafana, built in Cloud Combinator's account with sample data and a data-integrity test.
  • A costed migration plan — A migration plan for the entire environment reflecting the client's business and technical requirements, with an indicative monthly AWS run-rate of around $15.5k.
  • Handover — The solution was prepared for deployment into the client's AWS account, with a full demonstration and handover session.

With Bedrock assessed, the environment mapped, and a costed plan in hand, the client had a de-risked route into a full Azure-to-AWS migration. Cost figures above are indicative and depend on final workload patterns, region, and service configuration.

AWS Stack

Amazon Bedrock

For the agentic and generative AI inference plane powering content generation.

AWS Lambda

For the backend microservices and the agentic AI flow, deployed as containers.

Amazon ECR

For the container images backing the Lambda functions.

Amazon API Gateway

For REST endpoints equivalent to the existing Azure deployment.

Amazon RDS

For PostgreSQL for data storage, validated with a sample-data integrity test.

AWS Amplify

For building, deploying, and serving the frontend.

Amazon SQS

For asynchronous processing, and Amazon Managed Grafana for data dashboards.

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.