Platform Engineering (as Service) & Architecture - One Use Case

This has been marinating in my head for a while, it's time to spit it out. I will cover my approach at F****A around why, what, who and how I helped mature the organization. To give a brief background, when I landed there, I spent some time meeting with my peers, leaders and team to gauge the current state software architecture and delivery. I was also given a list of objectives from my leadership team, most of which were multi year scope, but I was able to start on the top few priority items immediately. I decided to jot these down so that I could spread this across the organization.

The What and Why?

This set the foundation of the capabilities and why they are important for the organization. 

Why Not?

I also decided to create some distinction of why we decided to move towards Platform Engineering from DevOps. It was important to capture this distinction for everyone. 

Who did What?

The next big question was how to structure the team around these capabilities (What). It's not practical to expect that everyone will be able to contribute in towards every capability. The cross pollination would happen as the team started to collaborate and mature over several quarters.  

 

How Team Delivered?

To keep these groups to a manageable size is to reuse them as much as possible, it was best to adopt "As a Service (PaaS)" mindset. This would allow the teams to work from a single backlog that combined strategic and tactical deliverables. It was time to take the SAFe recommendations (as R&D was following it closely) around the team structure and follow a more agile methodology (like Scrum or XP). 

Notice the Automation diamond above which was very intentionally from the start to move towards heavy automation and self service of the established and proven components. 

The OKR & Roadmap

One of the biggest challenge was how to distill the organization strategy into the team objective and roadmap. I used a good amount of thinking and collaborating in this area to create how the Platform Engineering would contribute to that organizational strategy and deliver results quarter over quarter. 

This was the resulting strategic roadmap developed with quarterly delivery model. 

Once we had this level of clarity, it was time to focus on how to show progress using KPIs and reports. We assembled the KPIs from Jira, Github, Environment and created reports that showcase whether these roadmaps were achieved. 

What about Architecture?

Though my position in this organization started as leader of Enterprise Architecture, I thought of it as a more of an enabler rather than a capability. Design and architecture was at the core of the platform and every capability needed a great deal of it, hence it was part of every capability and more. It also required a great deal of collaboration and focus from different parts of the organization. Hence, it made perfect sense to setup a light weight structure and process around it inspired by TOGAF, team topologies, SAFe, and agile methodologies. Here is how that looked like:

• Core Architecture Board - where core members and sponsors met periodically to groom the architecture backlog and made decisions.
• Working bodies - A small group formed to solve a small set of specific technology problem with POC and provide recommendation. 
• Implementation of these architectural decisions were left with one of the stake holders of the organization and was expected to be documented and operationalized via code/automation.

A few example of this initiative:

Where was GenAI?

As the process (PaaS) slides depicts automation and self service and due to resource constraint in developing in-house solutions, we leveraged the following solutions instead:

• Copilot - was enabled for the product development teams by migrating to Github cloud
• Virtual agents (MS Teams, Jira, Slack) were proposed to increase the efficiency of the teams based on the KPIs. 

Wrap up

The challenges were active participation of the software delivery and infrastructure teams even when their leadership were the supporters and sponsors of this structure. The platform team suffered tremendous amount of delay in delivering these key roadmap items as a result. 

Despite earning the best possible performance review, why am I in the job market? There's no one answer to this question. I reported to three supervisors during my 1.5 year journey at this organization and the last one being PMO leader from EU who didn't even know the team after six months. But, I was told that this team had achieved great deal in a short period but the new CIO organization didn't understand the value or the concepts of Platform Engineering & Architecture.