HomeCase Studies

Case studies

Client engagements on AWS

Each engagement summarises the business challenge, the approach taken, the outcome and the AWS services involved. Client identities are withheld under confidentiality agreements and identifying detail has been generalised.

Trusted by

Note on provenance. Engagements marked Client engagement are Rabbit Cloud delivery work; some are published with the client's name and others are anonymised under confidentiality agreements. Engagements marked Representative describe our standard delivery pattern for that workload type using illustrative figures and are not drawn from a single named client.

12 engagements

Software & SaaSClient engagement

TaskOPad · India

Production-grade AWS platform for a growing task management SaaS

Challenge. As customer numbers grew, the platform needed infrastructure that could scale with demand, isolate tenants reliably and release features without downtime, while keeping the operating cost appropriate for a growing software business.

Approach. Rabbit Cloud designed and built the production environment on AWS: a multi-account structure, containerised application tier with autoscaling, managed relational database with automated backups, CI/CD pipelines with environment promotion, and observability across the stack. Infrastructure was delivered as code so that environments are reproducible.

Outcome. TaskOPad operates on a scalable, monitored AWS platform with automated deployments and a run-rate aligned to usage, giving the team a foundation for continued product growth.

Amazon ECSAmazon RDSApplication Load BalancerAWS CodePipelineAmazon CloudWatchAWS IAM

Related service: DevOps & Platform Engineering Read the case study

Multi-tenantSaaS platform on AWS

Zero-downtimedeployment process

Construction TechnologyClient engagement

FollowUp CRM · United States

Reliable, secure AWS foundation for a construction-industry CRM

Challenge. A SaaS product serving contractors needs high availability during working hours, dependable data protection for customer pipeline information, and a release process that lets a small engineering team ship frequently without operational risk.

Approach. Rabbit Cloud established the production AWS environment with a highly available application tier, managed database with encryption at rest and automated backups, centralised logging and alerting, and a deployment pipeline with automated rollback. Security baselines were applied through AWS-native services from the outset.

Outcome. FollowUp CRM runs on an AWS foundation designed for availability and data protection, with an operational model that supports frequent releases from a small team.

Amazon EC2Amazon RDSAmazon S3Amazon CloudWatchAWS KMSAWS CodePipeline

Related service: DevOps & Platform Engineering

Multi-AZavailability design

Encryptedat rest and in transit

Construction TechnologyClient engagement

Sitemate · Australia

Smart construction data management on AWS

Challenge. Field data arrives in many forms and volumes, and must be captured, structured and made available for reporting and downstream product features while meeting the reliability and data-handling expectations of construction and infrastructure customers.

Approach. Rabbit Cloud supported the build-out of the data management layer on AWS: durable object storage for captured records and media, event-driven processing, a managed database tier, and the operational tooling to monitor throughput and cost. The design kept structured records available for both product features and analytics.

Outcome. Sitemate's data platform runs on AWS with durable storage, event-driven processing and the operational visibility required for a product handling growing volumes of site records.

Amazon S3AWS LambdaAmazon RDSAmazon EventBridgeAmazon CloudWatchAWS Glue

Related service: Data Engineering & Analytics Read the case study

Event-drivenprocessing pipeline

Lifecyclemanaged storage

HR TechnologyClient engagement

Superworks · India (ap-south-1)

Scalable, in-region AWS platform for an HR and payroll suite

Challenge. An HR and payroll product handles sensitive employee and salary data for many employers at once. The platform needed to scale with customer growth, keep data within India, and provide the security controls that HR and finance buyers expect.

Approach. Rabbit Cloud delivered the platform on AWS in the ap-south-1 Region: an autoscaling application tier, managed database with encryption and automated backups, network segmentation, centralised identity and access control, and a monitored, automated deployment process. The engagement is one of the larger SMB workloads Rabbit Cloud has taken to launch on AWS.

Outcome. Superworks operates a scalable, in-region AWS platform with the security controls appropriate to payroll data, supporting continued growth of the customer base.

Amazon EC2Amazon RDSAmazon VPCAWS KMSAWS IAMAmazon CloudWatch

Related service: Cloud Migration & Modernization Read the case study

ap-south-1data residency

Autoscalingapplication tier

PaymentsClient engagement

Paystone · Canada

Data management platform on AWS for a Canadian payments provider

Challenge. Payments and customer-engagement products generate data across several systems. The business needed a governed, secure data management layer on AWS that could support reporting and product analytics without duplicating data or weakening controls.

Approach. Rabbit Cloud supported the data management build on AWS: centralised object storage with lifecycle policies, catalogued and transformed datasets, a query and reporting layer, and access controls aligned to the sensitivity of payments data. Monitoring and cost visibility were included from the start.

Outcome. Paystone has a governed data management layer on AWS that serves reporting and analytics from a single, access-controlled source.

Amazon S3AWS GlueAmazon AthenaAWS Lake FormationAWS KMSAmazon CloudWatch

Related service: Data Engineering & Analytics

Governedsingle data store

Access-controlledby data sensitivity

Healthcare SaaSClient engagement · anonymised

Multi-tenant EHR and practice-management platform · United States

HIPAA-aligned AI documentation layer for a multi-tenant EHR platform

Challenge. Clinicians using the platform were spending a significant proportion of each day on documentation. The vendor required an AI capability, but multi-tenancy and PHI handling ruled out a general-purpose assistant.

Approach. Rabbit Cloud designed an encounter-to-note pipeline on Amazon Bedrock with Amazon Transcribe and Comprehend Medical, tenant-scoped retrieval on Amazon RDS for PostgreSQL with pgvector, and per-tenant AWS KMS keys. The cost model was established first so that the feature could be priced before development.

Outcome. A four-week proof of concept progressed to a production rollout plan, supported by an itemised production cost model in which AI and ML services represented the substantial majority of the workload.

Amazon BedrockTranscribe MedicalComprehend MedicalRDS PostgreSQLAWS LambdaAWS KMS

4 weeksproof of concept to first release

Per-tenantisolation and key management

Payments InfrastructureClient engagement · anonymised

Open-source payments foundation · Global

Adoption assistant for an open-source payments standard

Challenge. Implementers across multiple countries were raising the same integration questions repeatedly. Human support did not scale, and the documentation corpus was large, versioned and fragmented across sources.

Approach. A retrieval-augmented adoption assistant on Bedrock Knowledge Bases with Amazon OpenSearch Serverless, answering with citations into the specification and declining to answer where the corpus is silent. Positioned as ecosystem infrastructure rather than a vendor product.

Outcome. Implementers receive grounded, cited guidance at any hour, and maintainer effort is redirected from repeat questions to those requiring expert judgement.

Bedrock Knowledge BasesOpenSearch ServerlessAWS LambdaAWS Step FunctionsAWS Amplify

Related solution: Enterprise Knowledge Assistant

24x7self-service implementer support

Citedresponses traced to the specification

Healthcare ServicesClient engagement · anonymised

Patient access and pharmacy services organisation · United States

Document intelligence on AWS, integrated with an existing Azure estate

Challenge. An Azure-native organisation required AI document processing for patient access workflows. A full platform migration was neither required nor desirable, but the AI capability needed to operate on AWS.

Approach. Rather than proposing a platform migration, Rabbit Cloud scoped a net-new document intelligence layer on AWS using Amazon Textract and Amazon Bedrock, connected to the existing estate over private connectivity, with Amazon Macie monitoring the data landing zone.

Outcome. The organisation gained an AI capability without disruption to its core platform. The engagement produced a complete statement of work, cost model and target architecture for the production build.

Amazon TextractAmazon BedrockAWS Step FunctionsAmazon S3Amazon MacieAWS Direct Connect

Related solution: Document Intelligence Pipeline Read the case study

Multi-cloudno disruption to the existing estate

Net-newcapability without forced migration

Banking SoftwareClient engagement · anonymised

Core banking software ISV · India (ap-south-1)

Signature verification at branch scale for a core banking ISV

Challenge. Manual signature verification was a processing bottleneck across a large branch network, and any solution was required to retain customer data within India.

Approach. Computer vision models on Amazon SageMaker scoring specimens against reference signatures, Amazon Textract for instrument capture, Amazon ECS on Fargate for the service layer, and a reviewer console for scores within the uncertainty band. Deployed entirely within the ap-south-1 Region.

Outcome. A proof of concept supported by a fully itemised production cost model of more than forty line items, giving the ISV a defensible per-branch unit cost prior to commitment.

Amazon SageMakerAmazon TextractAmazon ECS on FargateAmazon S3AWS KMS

ap-south-1data residency by design

40+itemised production cost lines

Logistics & Supply ChainRepresentative

Regional freight and warehousing operator · Representative

VMware exit: 240-VM estate migrated to Amazon EKS and Aurora in fourteen weeks

Challenge. A hosting renewal and a licensing cost increase required a decision. The estate comprised a decade of accumulated virtual machines with incomplete documentation and no dependency map.

Approach. Discovery and dependency mapping preceded a 7R disposition for each workload. Approximately half were rehosted with AWS MGN, the transaction core was containerised onto Amazon EKS, databases were migrated to Amazon Aurora with AWS DMS, and eleven redundant workloads were decommissioned.

Outcome. Cutover was completed in five waves without unplanned downtime. Following rightsizing and migration to AWS Graviton, the run-rate settled materially below the hosting renewal quotation.

AWS MGNAWS DMSAmazon EKSAmazon AuroraAWS Transit GatewayAWS Control Tower

Related service: Cloud Migration & Modernization

240 VMsmigrated across five waves

14 weeksdiscovery to final cutover

Direct-to-Consumer RetailRepresentative

E-commerce and subscription retailer · Representative

One-third reduction in AWS run-rate with no impact on the product roadmap

Challenge. Cloud expenditure had grown faster than revenue for six consecutive quarters. Finance could not attribute cost to product lines, and engineering had no visibility of the cost of its decisions.

Approach. Allocation was addressed first through a tagging model with shared-cost treatment, giving every dollar an accountable owner. Rightsizing, migration of the stateless tier to AWS Graviton, S3 Intelligent-Tiering, non-production scheduling and a Savings Plans strategy sized to the new baseline followed in sequence.

Outcome. Run-rate reduced by approximately one third within a quarter with no feature work deferred, supported by a monthly FinOps cadence that has sustained the result.

AWS Compute OptimizerAWS GravitonSavings PlansS3 Intelligent-TieringCost and Usage ReportAmazon Aurora

Related service: FinOps & Cost Optimization

~33%run-rate reduction in one quarter

100%of spend attributed to an owner

B2B SaaSRepresentative

Vertical SaaS ISV · Representative

SOC 2 readiness and Well-Architected remediation delivered as one programme

Challenge. Enterprise sales were stalling in security review. The organisation faced a compliance deadline and an extensive findings register, with no capacity to address both independently.

Approach. A Well-Architected Review across all six pillars produced a scored findings register prioritised by exploitability. Rabbit Cloud remediated alongside the client's engineers: AWS Control Tower guardrails, IAM Identity Center, AWS Security Hub and AWS Config conformance packs, and automated evidence collection mapped to the control set.

Outcome. Findings closed with supporting evidence, an audit-ready control mapping, and a security questionnaire process that no longer delayed enterprise transactions.

AWS Control TowerAWS Security HubAWS ConfigIAM Identity CenterAmazon GuardDutyAmazon Macie

Related service: Cloud Security & Compliance

6 pillarsreviewed, scored and remediated

Automatedevidence collection

Get in touch

Request a detailed reference

Under a non-disclosure agreement, Rabbit Cloud can provide architecture detail, cost models and lessons learned from engagements relevant to your initiative.

Contact our team View case studies