Nike
Nike's apparel-samples process on AWS, provisioned entirely in Terraform with micro-frontends.
- Nike
- SMx Samples Management
- AWS and DevOps architect, full-stack developer
- 6 people

What it is
Sample products are ordered from factories, manufactured and shipped for sales, brand photoshoots, marketing and colour approval before production. SMx manages that cross-functional samples process.
The problem
Sample data arrives from external parties, several teams work on the same process, and infrastructure had to be reproducible across environments without manual console work.
Architecture
Requests flow left to right. Hover a component for its role.
- React micro-frontends client
- External sample data external
- Node.js Lambdas service, REST endpoints, event triggers
- ECS ingestion tasks worker
- SQS and SNS queue
- S3 data
- AWS Glue infrastructure
- Terraform infrastructure, state in S3
Key decisions
Everything in Terraform
The whole AWS footprint is defined in Terraform with remote state in S3, so environments are reproducible and changes are reviewed like code.
Ingestion and APIs scale separately
External data lands through ECS tasks, while REST endpoints and event triggers run on Lambda, connected by SQS and SNS.
Highlights
- Micro-frontend and microservice architecture across teams.
- ECS task-based ingestion for external sample data.
- Lambda for REST endpoints and event-driven triggers.