In response to a barrage of emails, to both personal and work email, asking if I am looking for a platform engineering certification (spoiler, I am not), I visited a website focused on platform engineering where I found a little surprise buried way down deep at the bottom of the page, just a small icon with no description.
It was a skull icon that linked to devopsisdead.com. DevOps is dead. Dot com. WTF!?
My mind’s eye was instantly taken back to a semi-traumatic event I experienced last year while doom scrolling on Instagram: A man screaming “DEVOPS IS DEAD” in an ad. I forgot what message followed; however, this is a particularly alarming position. As someone who has made DevOps his career, I can say definitively, DevOps is not dead.
I would be remiss to not explain my experience. I started working closely with engineers with the efficient delivery of software as my primary objective way back in 2008 while working at Playdom. This was around the same time Andrew Shafer and Patrick Debois met to discuss “Agile Infrastructure” at the 2008 Agile Conference. This is also about one year before the term “DevOps” became official in October of 2009 at the first DevOpsDays conference in Ghent, Belgium. I am an OG DevOps guy at heart, I understand the practices and principles at my core, and I feel proud ownership of the discipline as I was there, building it from scratch.
I’ve witnessed the evolution of DevOps practices firsthand. From leading cloud-native transformations with Kubernetes to building AI Infrastructure platforms, I’ve seen the consistency of only one thing: DevOps isn’t dead – it’s matured.
Platform engineering is an essential component of that maturity, not a DevOps replacement.
I would like to take the remainder of this article to offer my opinion on the impact and consequences of this stance; as well as to refute some points that have been made in its defense.
The State of the Conversation
The tech industry loves a dramatic headline. “DevOps is Dead” certainly delivers that shock value, generating clicks and conference attendance. But this framing creates an unnecessary and counterproductive divide in our community.
This messaging undermines the collaborative foundation that DevOps was built upon. By positioning platform engineering as DevOps’ replacement rather than its natural evolution, we risk alienating the very practitioners whose expertise built our current infrastructure landscape.
DevOps: Alive and Evolving
DevOps has always been about breaking down silos between development and operations to deliver software more efficiently. These core principles remain as relevant today as when they were first introduced. What’s changed isn’t the validity of these principles but the sophistication of their implementation.
Organizations from small startups to Fortune 500 enterprises continue to successfully implement DevOps practices at scale. The principles of automation, continuous integration/delivery and shared responsibility haven’t disappeared – they’ve become more refined and ingrained in how we work.
Platform Engineering’s Rightful Place
Platform engineering deserves recognition as a first-class citizen within the DevOps ecosystem. It represents a specialization that applies DevOps principles to create internal platforms that enhance developer productivity. Every DevOps organization I have managed has had a platform engineering function; it’s part of the complete DevOps breakfast.
A platform engineering team is essentially a DevOps team with a product mindset, focused on creating self-service capabilities for other engineering teams. This isn’t a replacement for DevOps – it’s a manifestation of DevOps principles applied to internal tooling.
The Reality of Tool Selection
One argument for the “death” of DevOps is the overwhelming number of tools in the CNCF landscape. This misses a crucial point: No organization is expected to implement all these tools.
Having built infrastructure across AWS, GCP, Azure, and various Kubernetes platforms throughout my career, I can confidently say that effective DevOps has always been about thoughtful tool selection. At every company I have been a part of in my career, and now at Lemurian Labs, I’ve led teams in selecting focused toolsets that serve specific organizational needs—never attempting to implement the entire CNCF landscape.
DevOps and platform engineers play a critical role in curating appropriate toolsets based on organizational requirements. The explosion of available tools actually highlights the importance of DevOps expertise in selection and integration, not its obsolescence.
The cognitive load argument falls flat when we recognize that tool selection has always been part of DevOps practice. Working with engineering teams across multiple organizations, I’ve seen firsthand that the proliferation of options doesn’t change our fundamental approach – it simply requires more disciplined curation.
Developer Experience: Always the Goal
Both DevOps and platform engineering share the goal of improving developer experience. The notion that platform engineering uniquely positions developers as “users” or stakeholders ignores that DevOps has always been about serving development teams more effectively.
Self-service capabilities aren’t new to platform engineering – they represent a continuation of DevOps automation principles. What platform engineering brings is a more formalized approach to packaging these capabilities, not a fundamentally new concept.
Adaptability at Scale
The claim that DevOps principles don’t scale effectively contradicts my direct experience leading infrastructure across organizations of various sizes and scale. Throughout my career, I’ve implemented DevOps practices that successfully adapted to organizational growth.
Platform engineering enhances this scalability by providing structured approaches to self-service. But suggesting that DevOps principles themselves don’t scale reveals a narrow understanding of those principles or their application.
Just because someone struggled to implement DevOps at scale doesn’t mean DevOps itself is flawed. Often, implementation challenges stem from organizational resistance or incomplete adoption, not from limitations in the DevOps approach itself.
The Future: AI and Automation
Interestingly, if any part of our stack is poised for AI-driven transformation, it’s the platform engineering portion. The standardization and tooling aspects of platform work lend themselves to automation in ways that the collaborative, cross-functional aspects of DevOps may not.
The core DevOps principles of collaboration, shared responsibility, and continuous improvement will remain essential even as implementation details evolve through automation. This further underscores that DevOps isn’t dying – it’s adapting.
Marketing vs. Reality
We should recognize provocative messaging like “DevOps is Dead” for what it often is – a marketing tactic designed to generate interest in platform engineering products and services.
Creating artificial divisions serves business interests that benefit from positioning their offerings as revolutionary rather than evolutionary. It’s far more compelling to market a solution to a “dead” practice than to acknowledge building upon established foundations.
This isn’t to diminish the value of platform engineering tools or approaches, but to place them in their proper context as enhancements to DevOps practices, not replacements.
Platform Engineering and DevOps: Better Together
Platform engineering deserves recognition as a critical component of modern DevOps organizations. Every DevOps organization should consider having dedicated platform engineering capabilities to build tools that enhance developer productivity.
But platform engineering uses DevOps practices and principles – it doesn’t replace them. The most successful organizations recognize this complementary relationship rather than forcing an artificial choice between the two.
As practitioners and leaders, we should resist divisive messaging that undermines the collaborative spirit at the heart of our community. DevOps isn’t dead – it’s deeply embedded in how we work, including how we approach platform engineering.
The future isn’t DevOps versus platform engineering. It’s DevOps and platform engineering, working together to deliver software more effectively than ever before.