DevOps is the status quo in product organisations. And not just for technology vendors, but now for enterprise IT as well. Now 80 percent of businesses are moving from multiple point solutions to DevOps platforms to streamline application delivery by 2027.
The drive towards large scale DevOps adoption is something I personally experienced as a software release manager for large scale development teams. My role was twofold – managing the Software Development Life Cycle (SDLC) and improving and streamlining the SDLC, so I always have automation of the DevOps process in the back of my mind.
DevOps adoption transforms how product organisations can deliver value in a few ways.
Of course, organisations benefit massively in their technical delivery. It is essential to ensure that developers have the necessary platforms to develop, test, collaborate on, and push changes from pre-production to production with automated controls at every level which are high quality and secure.
But what I also saw was how DevOps practices and an agile methodology could help enterprises not just transform the technical delivery of an application, but also create an environment where innovative ideas could be planned on an early and frequent basis, supporting teams with full backlogs. Moving faster from a technical standpoint enables and compels product organisations to think differently about planning and feedback.
Usage analysis, user interviews, and market data ensures that teams have insight into how they can improve features, and so drive future adoption. And when features are planned iteratively all considerations, acceptance criteria, standards, and pathways can be detailed iteratively, meaning you arm developers with a new robust backlog of requirements every sprint, not every quarter.
The role of AI in DevOps
Artificial intelligence (AI) can touch the SDLC at every stage to create much more streamlined and efficient processes. So, the DevOps promised land encompasses leveraging AI to analyse historical process mining data to uncover and prioritise the biggest inefficiencies and opportunities for product teams to go after.
AI can offload repetitive manual development tasks, so developers can work on projects which require more expertise and those which are complex. We have already seen a number of development efficiency gains in copilot capabilities. Here, AI is able to take a first pass at workflows, integration mapping and user experience components to name a few. This is expected to become more comprehensive and will continue to put knowledge at developers’ fingertips. In addition, there has already been a significant rise of AI-driven search, and this coupled with personalised AI tutors, developers of all levels can increase their productivity.
Where does planning fit in?
From countless meetings and workshops, plus drawing up reams of documents, and spreadsheets just to make a start in the planning process, it could feel impossible to be on the front foot. But the role of AI and generative AI here is transformative.
With generative AI, IT teams now have a discovery and planning assistant on hand that can convert legacy workflows into ones which are efficient and future-ready. The role of AI at all stages of planning can be broken down into a few key steps.
- Firstly, analysing historical operational process data, such as workflow diagrams and user manuals, to have a clearer idea of the current state of operations.
- Then, researching processes which involves looking through numerous industry best practices to understand what the possible approaches are and how to implement them into workflows.
- Finally, capturing all organisational considerations and goals from stakeholders and collating them into a much more coherent and all-encompassing vision to work towards.
As businesses seek to streamline their DevOps processes, AI and generative AI could be the spark that they need to get started. Through setting a foundational design for new initiatives, teams can start their development processes and collaborate much faster and easily. Over the past year, we have seen great progress made across the SDLC, even in those tougher to manage spaces, so we are getting much closer to the DevOps promised land.
Matt Healy
Matt Healy is Director of Intelligent Automation Strategy at Pegasystems.
