Log in to the AWS Management Console and search for the Elastic Container Service (ECS).
On the left navigation menu, select Task definitions.
Click the Create new task definition button.
Step 2: Configure Environment and Resources
In the Task definition family section, name your configuration (Example: ecommerce-ai-task).
Under Infrastructure requirements, select AWS Fargate as the compute platform.
Scroll down to the Task size section. Since the AI system running TensorFlow is resource-intensive, select 2 vCPU and 4 GB Memory to ensure uninterrupted processing.
Step 3: Configure Container and Port Mappings
Scroll down to the Container - 1 section.
Name: Enter a name for the container (Example: ai-backend-container).
Image URI: Paste the URI you copied from ECR in the previous step (Example: [Your-ID].dkr.ecr.[Region].amazonaws.com/your-ecommerce-backend:latest).
Under Port mappings, enter 3500 in the Container port field (This is the port we exposed in the Dockerfile). Leave the Protocol as TCP.
Step 4: Finalize Task Creation
Keep all other default environmental parameters.
Scroll to the very bottom and click Create. Your system blueprint is now ready to be deployed.