Case Study - Modernizing enterprise workloads with AWS and applied AI
TekStream engaged MarchForth as a delivery partner to accelerate a high-stakes cloud modernization program for a shared enterprise client.
- Client
- TekStream
- Service
- AWS architecture, AI integration, Platform delivery
Overview
TekStream worked with MarchForth to support a complex transformation program for a multi-entity healthcare services organization operating on aging on-prem infrastructure.
The engagement required both speed and precision. Legacy .NET and Java workloads had to be migrated with minimal downtime, while leadership needed better forecasting, more reliable operational insights, and a cleaner path to continuous delivery.
MarchForth joined TekStream's delivery team and owned key execution tracks end-to-end.
What we did
- AWS ECS and Lambda migration paths
- Amazon RDS and DynamoDB data tier strategy
- Event workflows with Amazon EventBridge and SQS
- AI support assistant using Amazon Bedrock
Delivery scenario
We designed a practical target architecture rooted in Amazon ECS Fargate for containerized services and Lambda for high-volume background workloads. Data processing pipelines were separated between Amazon RDS for transactional consistency and DynamoDB for high-volume event searches. EventBridge and SQS were introduced to break fragile integrations and mitigate failure amplification between services.
On top of the new cloud foundation, MarchForth built an internal AI triage assistant with Amazon Bedrock and OpenSearch-backed retrieval for runbooks and incident history. Operations teams used it to speed up diagnosis, recommend remediation paths, and surface relevant deployment context during active incidents.
The result was a stronger operating model delivered jointly with TekStream: faster release cycles, better observability, and lower cloud waste through workload right-sizing and improved deployment hygiene.