Has Agile lost its way? How AI-powered DevSecOps can help [Q&A]

Has Agile Lost Its Way? How AI-Powered DevSecOps Can Help

Twenty-five years after the Agile Manifesto set the stage for a revolution in software development, the movement that once promised responsiveness, customer-centricity, and empowered teams now finds itself at a crossroads. What began as a rallying cry for adaptive planning and collaboration has, for many organizations, devolved into a labyrinth of backlogs, burndown charts, and bloated frameworks. Teams across industries are increasingly voicing a familiar refrain: “We’re Agile, but…”—a subtle admission that something has gone awry. So, has Agile lost its way? And if so, what can be done to realign it with its original vision?

To explore these questions, we sat down with Bryan Ross, Field Chief Technology Officer at GitLab, who offers a compelling perspective: the future of Agile isn’t about discarding it, but about supercharging it with artificial intelligence (AI) and DevSecOps platforms to finally deliver on its founding principles.

The State of Agile Today: A Quarter-Century Later

When the Agile Manifesto was penned in 2001, its creators sought to address the rigidity and inefficiency of traditional waterfall methodologies. The manifesto’s emphasis on individuals and interactions, working software, customer collaboration, and responding to change struck a chord with developers worldwide. Over the years, Agile methodologies—Scrum, Kanban, Lean, and others—became the de facto standard for software development, promising faster delivery, higher quality, and greater customer satisfaction.

Yet, two and a half decades later, many teams find themselves mired in what Ross describes as “Agile bureaucracy.” Instead of empowering teams, the proliferation of processes, ceremonies, and tools has created new bottlenecks. Backlogs swell with unprioritized tasks, sprint planning becomes a chore, and retrospectives often feel like box-ticking exercises. The very frameworks designed to foster agility can, paradoxically, stifle it.

The “We’re Agile, But…” Syndrome

Ross points to a common sentiment among practitioners: the “we’re Agile, but…” mindset. This phrase typically precedes a litany of compromises—teams are Agile, but they’re still siloed; Agile, but they struggle with technical debt; Agile, but their deployments are slow and error-prone. These compromises signal a disconnect between Agile’s ideals and its real-world implementation.

One major culprit is the scaling of Agile practices. Frameworks like SAFe (Scaled Agile Framework) and LeSS (Large Scale Scrum) were developed to help large organizations coordinate multiple teams. However, critics argue that these frameworks often add layers of complexity, diluting the very principles they aim to uphold. Instead of fostering collaboration and adaptability, they can create new hierarchies and dependencies that slow decision-making.

Enter AI and DevSecOps: A New Hope for Agile?

According to Ross, the solution lies not in abandoning Agile, but in augmenting it with modern technologies—specifically, AI and DevSecOps. These tools, he argues, can help organizations reclaim the agility that Agile was meant to deliver.

AI as the Agile Enabler

AI has the potential to transform Agile practices by automating routine tasks, surfacing insights from data, and enabling faster, more informed decision-making. For example, AI-powered tools can analyze codebases to identify technical debt, predict the impact of changes, and even suggest optimal sprint plans based on team velocity and capacity. This frees up teams to focus on creative problem-solving and innovation, rather than getting bogged down in administrative overhead.

Moreover, AI can enhance customer collaboration by analyzing feedback and usage patterns to prioritize features that deliver the most value. This aligns with one of Agile’s core tenets: delivering value to customers early and often.

DevSecOps: Security and Operations at Agile Speed

DevSecOps—an extension of DevOps that integrates security and operations into the development lifecycle—offers another avenue for revitalizing Agile. By embedding security and operational considerations into every stage of development, DevSecOps enables teams to deliver secure, reliable software at the speed Agile promises.

Ross highlights that traditional Agile often treats security and operations as afterthoughts, leading to vulnerabilities and deployment delays. DevSecOps, powered by AI, can automate security testing, compliance checks, and infrastructure provisioning, ensuring that teams can move fast without breaking things.

The Path Forward: Realigning Agile with Its Principles

So, how can organizations harness AI and DevSecOps to restore Agile’s original promise? Ross outlines several strategies:

1. Automate the Mundane: Use AI to handle repetitive tasks like code reviews, test case generation, and backlog grooming. This allows teams to focus on high-value work.

2. Embrace Data-Driven Decisions: Leverage AI analytics to gain insights into team performance, customer needs, and technical risks. This supports more informed, agile decision-making.

3. Integrate Security and Operations Early: Adopt DevSecOps practices to ensure that security and reliability are built in from the start, not bolted on at the end.

4. Foster a Culture of Continuous Improvement: Use AI to identify bottlenecks and inefficiencies, and encourage teams to experiment and iterate based on data-driven feedback.

5. Keep the Human Element Central: While AI and automation can enhance Agile, Ross stresses that they should augment, not replace, human creativity and collaboration. The goal is to empower teams, not to create a “black box” development process.

The Future of Agile: Adaptive, Intelligent, and Secure

As we look ahead, it’s clear that Agile must evolve to meet the demands of modern software development. The rise of cloud-native architectures, microservices, and distributed teams has introduced new complexities that traditional Agile practices struggle to address. AI and DevSecOps offer a way to scale Agile without sacrificing its core values.

Ross envisions a future where Agile is not a rigid set of practices, but a flexible, intelligent framework that adapts to the needs of each team and project. In this future, AI handles the heavy lifting of planning and optimization, DevSecOps ensures security and reliability, and teams are free to innovate and deliver value at unprecedented speed.

Conclusion: Reclaiming the Agile Spirit

Has Agile lost its way? In many respects, yes—but only if we cling to outdated interpretations of what it means to be Agile. By embracing AI and DevSecOps, organizations can strip away the bureaucracy that has accumulated over the years and return to Agile’s roots: responsiveness, collaboration, and customer value.

As Bryan Ross puts it, the goal isn’t to replace Agile, but to realize its original promise. With the right tools and mindset, Agile can once again become the dynamic, empowering force it was always meant to be—only now, it’s smarter, faster, and more secure than ever before.

Tags: Agile, AI, DevSecOps, software development, Agile Manifesto, Bryan Ross, GitLab, automation, technical debt, sprint planning, customer value, collaboration, cloud-native, microservices, security, operations, continuous improvement, innovation, data-driven decisions, Agile bureaucracy, SAFe, LeSS, DevOps, software delivery, team empowerment, Agile evolution, modern development, Agile frameworks, Agile scaling, Agile challenges, Agile future, Agile transformation, Agile principles, Agile methodology, Agile practices, Agile teams, Agile tools, Agile processes, Agile ceremonies, Agile retrospectives, Agile backlogs, Agile burndowns, Agile compromise, Agile ideals, Agile implementation, Agile disconnect, Agile bureaucracy, Agile scaling, Agile frameworks, Agile evolution, Agile transformation, Agile principles, Agile methodology, Agile practices, Agile teams, Agile tools, Agile processes, Agile ceremonies, Agile retrospectives, Agile backlogs, Agile burndowns, Agile compromise, Agile ideals, Agile implementation, Agile disconnect, Agile bureaucracy, Agile scaling, Agile frameworks, Agile evolution, Agile transformation, Agile principles, Agile methodology, Agile practices, Agile teams, Agile tools, Agile processes, Agile ceremonies, Agile retrospectives, Agile backlogs, Agile burndowns, Agile compromise, Agile ideals, Agile implementation, Agile disconnect.

,