The Future of DevOps and the Rise of Intelligent Systems
Product & Delivery
In Part One of this series, we explored how DevOps emerged as a response to the challenges of modern software delivery and why many organisations have struggled to realise its full value. In this second article, Opencast Practice Lead Scott McCarthy examines how DevOps has matured beyond its original foundations and become a set of complementary practices focused on delivering reliable, sustainable outcomes at scale.
DevOps Today: Focused on Outcomes, Not Acceleration Alone
Modern DevOps has matured beyond the idea of speed at any cost. Today, it is about balancing delivery velocity with quality, reliability, and sustainability. The emphasis has shifted towards removing manual toil, improving feedback loops across the delivery lifecycle, and embedding quality into the flow of work rather than bolting it on at the end. Crucially, DevOps increasingly aligns with product thinking, recognising that delivery success is defined by user outcomes, not deployment frequency alone.
In practice, this means helping teams shift their focus from simply delivering features quickly, to understanding whether those features create meaningful outcomes for users. This is particularly important in public sector environments, where reliability, accessibility, and user trust are critical.
Organising Teams Around Value
One of the most important evolutions in DevOps thinking has been its alignment with product-centric team structures. Approaches such as Team Topologies reinforce the idea that teams should be organised around streams of value, supported by platforms and enabling capabilities. In this model, DevOps practices are embedded into delivery teams rather than owned by a separate function. Platform teams provide reusable, self-service capabilities, allowing teams to move quickly without sacrificing consistency or control.
As a Practice Lead, I’ve seen that teams aligned to clear value streams consistently outperform those organised around technical silos. When ownership is clear and teams are empowered end-to-end, delivery becomes faster, more predictable, and significantly less reliant on handovers.
The Evolution of DevOps: From Ways of Working to Intelligent Systems
Modern DevOps has evolved far beyond automation and CI/CD (Continuous Integration and Continuous Deployment/Delivery) pipelines. Today, it encompasses Build and Release, Platform Engineering, SRE (Site Reliability Engineering), DevSecOps (Development, Security, and Operations), FinOps (Financial Operations), MLOps (Machine Learning Operations), and increasingly AI-driven operations.
As systems and organisations have grown in complexity, DevOps has naturally evolved into a set of more focused disciplines to address specific concerns such as security, cost management, reliability, and AI model lifecycle management. These are not deviations from DevOps, but a reflection of its continued maturation.
Specialisation isn’t the opposite of DevOps — it’s a sign that DevOps has matured.
Reliability as a First-Class Engineering Concern
SRE has had a particularly strong influence on modern DevOps. By treating reliability as an engineering problem, SRE introduced concepts such as service-level objectives, error budgets, and blameless post-incident reviews. These ideas helped organisations move away from fear-based operational models towards informed, data-driven decision-making. SRE doesn’t replace DevOps; it provides concrete mechanisms for delivering on DevOps’ promise at scale.
Platform Engineering: DevOps That Scales
As organisations continued to grow, many struggled to apply DevOps practices consistently across teams. Platform Engineering (PLE) emerged as a response, focused on creating internal platforms that offer standardised, self-service capabilities.
Rather than asking every team to solve the same problems independently, platform engineering turns DevOps practices into consumable services. This reduces cognitive load for delivery teams while improving security, consistency, and developer experience.
In my experience this approach has proven particularly effective in scaling delivery across multiple teams, where standardisation and self-service can reduce duplication while still enabling autonomy.
Build & Release Engineers fill the gap between SRE and PLE disciplines, aligning closely with the original DevOps focus of building and releasing software. These engineers work within software delivery teams to ensure efficient releases in large-scale public sector environments. Google's 2016 SRE Handbook defined the distinct role of Build & Release Engineers (BRE) versus Site Reliability Engineers (SRE). Many organisations still refer to DevOps as BREs but often lack the maturity to distinguish DevOps into specialized engineering roles.
AI and the Next Evolution of DevOps
AI is now shaping the next phase of DevOps evolution. AI is already influencing how code is written, tested, deployed, monitored, and secured. In operations, AIOps (AI for IT Operations) is enabling teams to move from reactive monitoring to predictive and automated responses. As systems become more intelligent and autonomous, DevOps will continue to evolve — shifting from manual control to guiding, governing, and collaborating with AI-driven systems.
However, adopting AI effectively will require the same foundations that DevOps has always relied on — strong engineering practices, clear ownership, and a culture of continuous learning. Without these, AI risks amplifying existing inefficiencies rather than solving them.
AI is still in its infancy, but already it has introduced new tooling and ways of working such as:
Model Context Protocol (MCP): is an open-source standard that acts as a universal bridge for AI applications, often called the "USB-C for AI." Developed by Anthropic and hosted by the Linux Foundation, it standardizes how AI models securely connect to external tools, databases, and code repositories.
Language Server Protocol (LSP): Originally created by Microsoft, it is a standardized, machine-oriented communication protocol used between development tools (like IDEs or AI agents) and "language servers" (like Pyright for Python or gopls for Go). It decouples the code editor from the language analysis tools.
Engineering Intelligence Platform (EIP): is a centralised analytics tool that brings together data from across the software development lifecycle. It helps leaders spot waste, gauge productivity, and connect technical performance to business goals.
Key features include:
Data Integration: Consolidates data from Jira, GitHub, CI/CD, and incident tools into one dashboard.
Delivery Metrics: Tracks engineering standards like DORA metrics automatically.
Developer Experience: Measures developer friction and well-being using activity and survey data.
AI Tool Impact: Monitors how AI coding assistants affect productivity.
Resource Allocation: Maps hours and costs to features for better budgeting (including AI credits usage).
The future of DevOps is not just automated — it’s increasingly intelligent.
Looking Ahead
DevOps has never stood still. From improving collaboration, to embracing cloud-native delivery, to enabling platform engineering and AI-driven operations, it continues to evolve alongside technology itself. The organisations that succeed will be those that remember DevOps’ original purpose: enabling people to deliver value, learn quickly, and build systems that can adapt to constant change.
Just as DevOps transformed the tech landscape, software vendors are now incorporating AI into their products to enhance value. However, this influx of AI features can create additional confusion. While AI-enabled tools offer significant efficiency gains, they may also overwhelm engineers working in the field with too much information. Remember 3D TVs? Unlike them, I believe AI is here to stay. When used responsibly and ethically, especially by skilled engineers, AI can truly be a ‘force multiplier’ that boosts productivity.
The tools will change. The terminology will evolve. But the challenge remains the same. Deliver better outcomes — reliably, sustainably, and together.
Final Thoughts: An Evolved DevOps Approach at Opencast
The future of DevOps is unlikely to be defined by a single team, role, or toolset. Instead, it will be characterised by intelligent systems, product-focused delivery teams, platform capabilities, and continuous learning. DevOps engineers will need to actively embrace AI, just to keep up with the pace of change. The organisations that thrive will be those that view DevOps not as a destination, but as an evolving discipline that continually adapts to new technologies and changing business needs.
At Opencast, based on our experience supporting complex delivery environments, we continue to believe that strong DevOps foundations are critical to successful modern software delivery. However, we also recognise that DevOps has evolved significantly since its original inception and can no longer exist in isolation. This is why we go beyond traditional DevOps methodologies alone.
Our ways of working intentionally align DevOps practices with a strong product mindset, informed by user-centred design and underpinned by modern software, ensures that delivery teams are focused not only on speed, but on creating meaningful, sustainable outcomes for users. This integrated mindset enables our software engineering teams to perform at their best, empowered to deliver high-quality, secure, and resilient solutions, while remaining adaptable in an increasingly complex and fast-moving technology landscape.
OpenPerspectives is our platform for Opencast people to share their thoughts and perspectives on modern digital delivery. It offers practical insight into user-centred design, engineering excellence, product leadership, data-driven decision making and building expert capabilities, grounded in real-world experience.
