The platform teams are focused on developing and maintaining reusable services for their customers– the individuals in the different value streams, across the organization. The sole success criteria for the platform teams should be the ease of adoption, usage, and documentation of the services they provide. The more compelling and easier to use these services are, the wider the adoption.
In Chapter 1, Accelerating Your DevOps Journey with AWS, we discussed the idea of reducing cognitive load for customers. The concept of a platform team strongly relates to this idea. They should enable the value stream members to be able to focus on differentiating business outcomes. This is only possible by offloading other undifferentiated tasks from them. The outcomes from the platform team address the common needs of all other value streams. For example, if all the software artifacts depend on containerized workloads running Amazon Elastic Kubernetes Service (Amazon EKS), then the right focus area for the platform team should be ready-made templates for deploying, monitoring, and operating services in these EKS environments. They could offer centralized dashboards and out-of-the-box logging integrations for the value streams to easily adopt and benefit from.
However, it is important to note that ownership of the production environments of software developed by respective value streams still sits with them.
In Chapter 11, Ensuring a Strong AWS Foundation for Multi-Account and Multi-Region Environments, we will discuss the concept of a Landing Zone. This maps well to the scope of work that a platform team would ideally deal with, in the context of AWS environments. This enables the application teams to focus on what they do best while leaving the complexities around the foundational aspects of a multi-account, multi-region environment with the central platform team.
Specialist teams
Depending on an organization’s domain of business, there might be certain specialized areas that are overarching across most, if not all value streams. These areas require special skills that are not commonly found in the industry, or within the company. This could be deep expertise in media encoding, financial risk analysis algorithms, high data compression techniques, and so on.
In all likelihood, this knowledge is core to the artifacts that the value streams provide. It is essential to continue providing value to the end customers. In such cases, the organizations can consider establishing these specialist teams who consult the value streams on best practices around the niche skillset. Highly mature DevOps companies also invest in offering these specialized skills as a service, often abstracting the underlying details with easy-to-understand interfaces, thereby reducing the cognitive load of the value streams.