DEVOPSdigest

DEVOPSdigest invited experts across the industry — consultants, analysts and vendors — to comment on how AI can support the software development life cycle (SDLC). In Part 16 of this series, experts offer predictions about how AI will change DevOps and development processes in 2025 and beyond.

FUSION OF AI AND DEVOPS

AI and DevOps: The Predictive Production Line:
In 2025, AI will supercharge DevOps by predicting bottlenecks and preemptively suggesting optimizations, transforming DevOps pipelines into "predictive production lines." With Gartner forecasting that AI will drive nearly 80% of emerging tech investments by 2025, the fusion of AI and DevOps will create workflows that fix issues before they impact production.
Ravi Ithal
GVP and CTO of Proofpoint DSPM, Normalyze(link is external)

A RANGE OF TASKS ACROSS THE DEVOPS PIPELINE

DevOps teams are increasingly exploring how GenAI can be used in the DevOps pipeline. As per IDC: 50.1% of respondents are expanding, using or piloting generative AI in their DevOps pipeline, demonstrating how fast GenAI is being adopted by the DevOps teams. An additional 21.9% are planning to adopt it within a year. We're seeing GenAI used commonly across intelligent testing, chatbots and natural language interfaces and knowledge management.
Prashanth Nanjundappa
VP, Product Development, Progress(link is external)

Sci-fi films predict the sophistication of AI, and, to an extent, its level of sophistication will continue to evolve. AI models will become capable of generating more complex code and handling a wider range of tasks. We'll see demand for specialized AI tools tailored to specific and niche development domains (e.g., web development, mobile app development, data science) or programming languages.
Dotan Nahum
Head of Developer-First Security, Check Point Software Technologies(link is external)

In 2025, AI-driven automation will go beyond simple task efficiencies to reshape entire DevOps pipelines. With the ability to dynamically adapt to workload shifts, AI will help teams automate complex workflows and reduce manual oversight, creating a balance that frees developers to focus on innovation while maintaining quality control over deployments.
Fitz Nowlan
VP of AI and Architecture, SmartBear(link is external)

MULTI-STEP END-TO-END PROCESSES

Developers have moved beyond the fear that AI is replacing what they do, and instead expect it to augment their work. This has made them more comfortable incorporating AI tools into their daily workflows. Next, we'll see these tools applied beyond isolated tasks and use cases to multi-step processes across entire software development life cycles. This evolution will be further accelerated by advancements in AI capabilities in the coming year. Companies are making AI tools more accessible, reliable and capable of handling a broader range of development tasks based on what they've learned from initial insights and applications. And as we've seen in other use cases, AI-enabled agents have begun to take center stage. As this happens, we'll see AI taking on more end-to-end responsibilities, beginning as early as initial planning and design to deployment and ongoing maintenance. These responsibilities will also become more sophisticated, including everything from tackling complex coding tasks to automating testing, and even taking actions on behalf of development teams in managing entire segments of the software development life cycle.
Tom Cozzolino
Chief Strategy Officer, USA, GFT Technologies(link is external)

AUTONOMOUS AI AGENTS

DevOps will be transformed by agentic AI to be more highly automated. This will necessitate an increased level of planning and more complete, unambiguous requirements so that the DevOps agents will be able to build the right thing.
David Brooks
SVP of Evangelism, Copado(link is external)

In the future, AI code assistants have the potential to revolutionize technical processes and substantially increase the speed of product delivery. In addition to empowering developers by giving them time to focus on more demanding tasks, these tools will increasingly enable AI "agents" that can autonomously complete tasks without relying as much on technical teams for analyzing requirements, identifying problems, and fixing and generating code. These capabilities will help to further democratize the creation of new products and features, and organizations will benefit from faster returns on their investments.
Keri Olson
VP of Product Management, AI for Code, IBM(link is external)

AI AGENTS START IN OPEN-SOURCE ECOSYSTEMS

AI Agents will be catalysts for software supply chain transformation. AI agents are poised to revolutionize the software supply chain by automating and optimizing processes, from continuous integration to continuous deployment. This transformative shift will initially gain traction in open-source ecosystems, where AI agents are likely to be built and shared with the community, like software libraries. As developers and organizations witness the benefits of AI-driven automation in open-source projects, we can expect a rapid expansion into commercial enterprise solutions. Internal development teams and platform engineers will increasingly be tasked with building, extending, and integrating AI agents across the entire software supply chain.
Lee Faus
Global Field CTO, GitLab(link is external)

THE TROUGH OF DISILLUSIONMMENT

GenAI will hit the trough of disillusionment around the end of the year, but to get there, many teams will experiment with applying GenAI solutions to various points in the DevOps toolchain with varying degrees of success. The generation of project documents will have some success, but there will be problems with code snippet generation.
Helen Beal
CEO and Chair, Value Stream Management Consortium(link is external)

PLATFORM ENGINEERING

AI will drive efficiencies for platform engineers. The proliferation of pattern recognition in AI technologies is expected to reduce the friction of automating software releases into production. By creating reusable building blocks that encapsulate common functionalities for software delivery, platform engineers will help empower non-technical team members to easily assemble delivery pipelines using intuitive low-code techniques for testing, environment management, and release orchestration. This movement will lead to a rise in application development driven by AI-assisted tools, enabling organizations to meet specific needs more efficiently.
Lee Faus
Global Field CTO, GitLab(link is external)

SBOMS

AI-based automation bots that analyze deep software bills of material (SBOMs) will simplify complex software maintenance decisions and automate remediation. This innovation is expected to reduce effort spent on software maintenance and cut software upgrade costs.
Javed Hasan
CEO and Co-Founder, Lineaje(link is external)

PLANNING AND PROTOTYPING

The biggest change will be in the use of AI for planning and rapid prototyping. This will impact the quality of the user experience. Most UX designers know how to make great experiences, but most teams are constrained by schedule and often compromise to hit a deadline. GenAI will make it possible to generate UI in minutes with no compromises.
David Brooks
SVP of Evangelism, Copado(link is external)

USER INTERFACES

Multimodal LLMs, such as OpenAI GPT 4o, will open up entirely new sets of use-cases for software development, particularly for GUI applications. Think about how you interact with applications in your day-to-day life. Applications are built around the user interface and being able to send the same "input" we experience (e.g. a screenshot or video of the UI) to an LLM opens up many possibilities for how AI can more accurately interact with applications in the same way that humans do.
Todd McNeal
Director of Product Management, SmartBear(link is external)

API DESIGNS

AI-generated API specifications will transform the API design process by automatically creating comprehensive, standardized specifications based on business requirements. These systems will understand best practices, security requirements, and backward compatibility needs, ensuring consistent API design across organizations while reducing development time.
Tristan Stahnke
Principal Application Security Consultant, GuidePoint Security(link is external)

CI/CD PIPELINES

We can expect greater integration of AI in continuous integration/continuous deployment (CI/CD) pipelines, enhancing automation and efficiency.
Jabari Holder
AI Principal & GenAI GTM Lead, AND Digital(link is external)

AI IN MAINFRAME ENVIRONMENTS

2025 will mark a critical inflection point where AI-augmented development tools become essential bridges across the growing experience gap in enterprise IT, particularly in mainframe environments. Looking at the respondents for the 2024 BMC Mainframe Survey, veteran mainframe experts have declined dramatically from 36% to 13% since 2019 as they retire, while the number of emerging mid-career professionals has risen to 36%, or more than a third of the respondents. In this shifting landscape, AI-powered development assistants will serve as virtual mentors, encoding decades of best practices and institutional knowledge to support the rising generation of practitioners. However, organizations must carefully balance this AI augmentation with robust testing frameworks and quality controls to ensure that increased development velocity doesn't come at the cost of system reliability — especially as less experienced developers become more reliant on AI-generated solutions for complex mainframe operations.
John McKenny
SVP and GM of Intelligent Z Optimization and Transformation, BMC Software(link is external)

LLM ORCHESTRATION ENGINES

Watch for dev teams to increasingly leverage LLM orchestration engines to make the most of available technologies in the coming year, while providing more finely customized results for application users within their organizations.
Shomron Jacob
Head of Applied Machine Learning & Platform, Iterate.ai(link is external)

MANAGING AI PROCESSES

While AI implementations may at first take shape by using the technology to tackle more complex coding and other DevOps tasks, we'll eventually see companies using AI to manage their other AI processes. For example, developer teams will no longer have to move from system to system to access the information and tools they need. They'll instead interface with these dedicated AI agents that will work with all of the other systems on their behalf, significantly streamlining workflows and productivity. Rather than simply using AI tools, organizations will begin transforming the way they operate.
Marco Santos
Co-CEO, GFT Technologies(link is external)

Go to: Exploring the Power of AI in Software Development - Part 17: 2025 Predictions and Beyond