Platform engineering is emerging as the key to scaling DevOps practices through self-service capabilities, according to Andi Grabner, Platform Engineering Advocate at Dynatrace. The focus is on delivering the right data to the right people at the right time.
At Dynatrace Perform 2025, the company introduced Observability for Developers, providing development teams with easy data access, advanced analytics and the new Live Debugger. This supports what Grabner calls the “one-click opportunity” – integrating debugging and observability tools directly within developer workflows.
“The key consideration is what type of observability data is needed for which consumer of the data,” Grabner explains. “I need to understand where the developer spends most of their time in tools and then figure out how to get the relevant observability data to the developers where they are when they need the data.”
Implementation requires careful consideration of developer workflows. “Backstage or any other developer portal is a good place to provide a high-level overview to the developers,” Grabner notes. “The consideration is what observability data needs to be pushed when in the development process, into which tools so that the developer gets it when they need it, without having to ask for another context switch.”
According to Mitch Ashley, VP and Practice Lead, DevOps and Application Development, The Futurum Group, “Our research shows continued strong adoption of both DevOps and platform engineering. Tensions between DevOps and platform engineering teams are short-lived. Platform engineering plays a vital role in developer productivity, increasing flow and simplifying many complexities.
The platform’s enhanced Davis AI capabilities now support preventive operations, which Grabner sees as crucial for platform teams. “Preventive operations is a great way to make sure that your platform and also the services that you deploy to the platform are properly sized when resizing is necessary,” he explains. The AI engine can predict scaling requirements and automatically generate recommended configuration changes through pull requests.
Templates are key to scaling these capabilities effectively. “These templates should include all the best practices we have seen currently in the CNCF for proper configuration and specification of metadata on your deployment,” Grabner notes. These templates incorporate both industry standards and organization-specific requirements.
“How do you scale this? Well, the scaling comes implicitly because templates mean people can use the templates,” Grabner continues. “And then if you have 1000 teams, they would use the same template.” The challenge lies in managing updates: “The only thing you need to figure out as an organization is when the template changes because there might be a new regulation coming in, how do you propagate that template change to the existing software services?”
These capabilities align with Dynatrace’s vision of self-service implementation for enterprise-wide adoption. The platform enables teams to provide tailored entry points and integrations with developer portals and IDEs while maintaining compliance through automated checks and fine-grained data segmentation. As Grabner emphasizes, “Observability is no longer optional. Observability is mandatory. Platform engineering must enable observability for everything you do on your platform; otherwise, you’re flying blind.”
The future of platform engineering lies in this combination of AI-driven automation and self-service capabilities. While human oversight remains crucial, platforms can increasingly suggest optimizations and improvements automatically, helping organizations scale best practices effectively.