
A large investment firm was acquired and had to move its AWS platform onto the new owner’s AWS environment. The migration was constrained by the size of the platform, the applications running on it, and EU and US regulations the new owner had to follow.
Workloads moved between AWS environments with controlled downtime and a documented rollback path.
Cloud infrastructure managed centrally, with unified governance across the new environment.
AWS CDK keeps the new infrastructure version-controlled and repeatable.
Lambda, API Gateway and serverless patterns where they fit, to keep the integration small.
Event-driven data flows so payments and other in-flight data kept moving.
Python, Node.js and TypeScript across the data transfer applications.
Our customer, a large investment firm, was acquired by a competitor. Their AWS platform had to migrate and integrate into the new owner’s AWS environment. With the new owner based in the United States, we had to satisfy both European and US regulations.
Meanwhile, the existing platform had to keep running, including the data flows and payment services.
The migration ran in several phases. The first approach was to use an internal tool the new owner had built to onboard new AWS accounts and applications.
We then looked at the most critical services: the databases. The data had to move from the old platform to the new one, secure and compliant with EU and US rules. The internal tooling did not support the restoration method we needed.
Given the regulations in financial services on both sides of the Atlantic, we needed a different approach to organising the new accounts and applications, while payment services and data flows kept running.
Another company had been acquired by the same owner some time earlier and had completed a similar migration. They were willing to share what they had built: a Terraform-based setup that fit our case as well, and we adopted it as the base for the new platform.
Payment flows needed a custom solution. Data was extracted from the old platform — sometimes directly from the database, sometimes from an application running on Kubernetes — and dropped into an S3 bucket. A Lambda function processed and enriched it, then wrote it to another S3 bucket. From there, a secure data transfer service moved the data to the new owner’s platform.
The data transfer application was built with the AWS Cloud Development Kit (CDK), so the infrastructure stayed in code and the deployments were repeatable. The Lambda functions and the application code use a mix of Python, Node.js and TypeScript. The architecture is event-driven: data is processed in small chunks, with enrichment and validation along the way.
The Factory has migrated many workloads, including cloud-to-cloud moves between AWS accounts and organisations. We help define the migration approach, keep data flows running during cut-over, and rebuild the new platform with Infrastructure as Code so deployments stay consistent and repeatable.
Read the service