25 years after the Agile Manifesto codified the habits of high-performing teams, the methodology remains the bedrock of software engineering. It successfully dismantled the monolithic planning structures of the past, but the current uneven adoption of AI threatens to destabilise established workflows.
Martin Reynolds, Field CTO at Harness, argues that while the tools change, the core objective remains constant. “It’s difficult to overstate how much the Agile Manifesto has reshaped software delivery over the past 25 years,” he says.
“By codifying what the most effective teams were already doing, the Agile Manifesto helped move the industry away from rigid waterfall models with long planning phases and infrequent releases toward a more sustainable approach to driving innovation.”
The velocity trap
Agile worked because it compressed the timeline between committing code and verifying execution.
“Work shifted into shorter iterations with fast feedback built into CI/CD pipelines,” Reynolds explains. “Teams could catch defects earlier, deploy incremental improvements safely, and see the impact of their work—improving morale and ownership.”
This approach paved the way for modern architecture. The Agile Manifesto “laid the groundwork for today’s cloud-native world, where ephemeral infrastructure and dynamic deployments reflect the way teams actually work.”
The friction arises when AI enters this ecosystem asymmetrically. Developers currently use Large Language Models (LLMs) primarily to accelerate code authoring. When the “write” phase speeds up without a corresponding acceleration in the “verify” phase, the delivery pipeline creates a bottleneck.
“AI-generated code is already being used to speed up development at the front end, but those gains are often offset downstream, creating more bugs, higher cloud costs, and greater security exposure,” Reynolds warns.
For engineering leads, this manifests as increased failure rates in staging and higher cloud bills due to inefficient machine-generated queries running on auto-scaling infrastructure. The technical debt accumulates faster than manual code reviews or static analysis tools can mitigate.
Balancing the equation
Restoring equilibrium requires applying intelligence to the validation process, not just creation. “The real value will emerge when AI is extended beyond code creation into testing, quality assurance, and deployment,” Reynolds notes.
Implementation involves moving beyond basic automation to predictive orchestration. Platform engineers can implement ML-based test selection, where the system identifies and runs only the tests relevant to a specific code change, rather than a full regression suite. Similarly, AI-driven anomaly detection in production logs can identify performance degradation often missed by static thresholds.
“Applied thoughtfully, it strengthens the fast feedback loops that Agile depends on and helps teams adapt more confidently as business requirements change,” says Reynolds.
Evolution of the Agile Manifesto
The influx of generative tools does not render iterative development obsolete; it increases the necessity for robust feedback loops.
“Looking ahead, AI presents a real opportunity for Agile to evolve into its next phase—but only when applied across the full delivery lifecycle,” explains Reynolds.
For architects and CTOs, the strategy involves looking past developer productivity metrics (that are often a vanity number) and focusing on system throughput and stability.
“In that sense, AI doesn’t replace or outdate Agile—it reinforces the principles that have kept it relevant for the past quarter of a century and will ensure it remains so long into the future.”
See also: Cyber Security Expo 2026: Machine trust in modern software delivery
Want to learn more about cybersecurity from industry leaders? Check out Cyber Security & Cloud Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the AI & Big Data Expo. Click here for more information.
Developer is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.