Q-Bits :Starting with the architecture and use cases, utilize Amazon Q developer to build the necessary infrastructure
Welcome to another installment of Q-Bits, our regular series showcasing cool ways Amazon employees are leveraging Amazon Q Developer. In this article, we will demonstrate how to start with your architecture and use case requirements and utilize Amazon Q Developer to automatically generate the necessary infrastructure.
Abhijit Rajeshirke
Amazon Employee
Published Mar 25, 2025
In this article, I will demonstrate how you can start with your desired architecture or use case, and then leverage Amazon Q developer to help build the necessary infrastructure.
Recently, I was developing a workshop to show customers how they can use Amazon Athena's IP function to identify VPC traffic by querying CIDR blocks. To enable this use case, I aimed to build an architecture similar to the one shown below.
For of usecase I wanted to have Three Virtual Private Clouds (VPCs) to be created: VPCA, VPCB, and VPCC. VPCA and VPCB are connected using VPC peering, allowing resources in the two VPCs to communicate with each other.
So now, I have high level requirement outlined, I will refine the requirements to be more specific, including details like the CIDR blocks needed for the VPCs, the number and types of subnets required, and other precise specifications.
VPCA and VPCC are connected using a Transit Gateway, which enables communication between the resources in the two VPCs. Additionally, VPC flow logs have been enabled for VPCA to capture information about IP traffic going into and out of VPCA subnets. The flow log data is stored in an S3 bucket for analysis.
The aim is to provide comprehensive and exact requirements upfront so there is less need for back-and-forth adjustments with Amazon Q Starts recommendations. However, the process can still be iterative – and you can progressively update the requirements and Amazon Q will adapt its output accordingly. The Amazon Q allows flexibility to refine the specifications as needed while generating customized cloud architecture recommendations.
Here is Final Prompt :
Amazon Q Developer streamlines much of the process of creating CloudFormation templates. However, since Generative AI is not deterministic, Q Developer may produce slightly different results each time it is used. Some manual adjustment of the generated CloudFormation templates may be necessary to fully align them with your specific needs.
Here is final cloudformation template .
Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.