DevOps Plus 2025 – Predict 2025

Hello everyone. I am here to talk about state of art software development in 2030 and beyond. Have you ever thought about what software development look like in 2030?

How AI will participate in shaping the future of software engineering, and what will be our major problems or challenges by then? What will be the visionary ideas for overcoming those problems and challenges? Obviously, I will not have all the answers today, but I'm going to talk about DevOps plus for 2025, which could answer some of these questions and keep you ahead of the technology curve.

Welcome to Predict 2025. If you're interested in my talk. I am Garima bpe, founder for the Canada DevOps Community of Practice, producers of Summit Canada Chair for the Ambassador program at Condensed Delivery Foundation.

I was awarded DevOps Executive of the Year by a DevOps dozen last year. My two progressive books, um, strategizing Condensed Delivery in Cloud and CICD Design patterns. My call to action is leadership practices and communities, and if you wanna connect with me, my LinkedIn handle as well as my communication email is at the bottom of the slide.

So if you have been following me on social media, I often start my talks with very basic fundamental understanding about the topic. So let's, uh, go back in time, a decade back. And why DevOps?

What are the core values of DevOps and why did the teams, organizations, and leaders lean upon DevOps? Uh, so obviously there, uh, there has a lot has happened during one decade, but the key, you know, outline or values remain the same. Um, DevOps emphasizes collaboration, building a culture that un unlocks the potential of the teams through collaboration and harnessing relationships to maximize business outcome.

And in the context of in the era of ai. Now, we would also have to build collaboration with human machine, uh, uh, engagements. Lean, considering lean practices to eliminate waste and add efficiency, avoid overproduction context switching, eliminate management overhead.

And this all was a I byd in the principles of DevOps, sharing, setting clear goals, uh, making it shareable, preparing the teams, breaking down the lateral, horizontal and hierarchical barriers, and developing a goal oriented cross-functional approach, which naturally brings alignment. DevOps also fostered automation by understanding the critical flow, complementing it with tooling and upskilling and modeling integrity into the system while becoming more transparent and consistent. And lastly, I would say measurement mechanism to communicate and analyze how improvements have helped achieve business outcome and supporting cultivation of best practices.

And obviously, um, in a decade's time, we have, um, come a long way. We have accomplished a lot, but now, uh, in 2025, I would advocate the DevOps strategy needs of facelift. And during my talk today, we will actually gather some inputs and some insights on the facelift and the market outlook of the space facelift.

So, uh, starting from the basics, 74% of the organizations already have, um, been implementing DevOps in some capacity as reported by puppet lab. 01 billion by 2026 according to markets and markets. DevOps market size has been projected to grow by 20%, CGR from 2023 to 2032, driven by the rising need of reducing software development cycle and accelerating dev, um, delivery of software.

Obviously, there is one other aspect, which is everybody's talking about AI and, you know, riding on the top of AI wave. So $1 trillion, uh, in development of AI initiatives in the next five years, and how DevOps can foster that strategic initiative through, uh, tooling application through the capability which we have, uh, inculcated in a decade's time. So I'll talk about a little bit, uh, more about, you know, how the trends are looking like, what kind of, you know, facelift, uh, we are looking at 2025 and what are strategic objectives as DevOp practitioners and leaders in the next few slides, but starting from our favorite Dora report and applying insights from dora.

So what Dora is suggesting, uh, from 2024 report is, and some of these, um, findings are, uh, surprising, like AI is hurting delivery performance, and, um, it reflected that effect on delivery. Throughput is small, but likely negative. The ne negative impact of delivery stability is larger.

So there is, uh, some, you know, work which needs to be done in terms of how we join hands with this AI movement and how do we, um, integrate AI learnings and AI assistant technology into DevOps. Another aspect is platform engineering. Uh, through this report, uh, Dora emphasized that internal development platform users had 8% higher levels of individual productivity and 10% higher levels of team performance.

Moreover, organizations software delivery and operation performance increased by 6% when using a platform. So obviously platform engineering is a clear winner through the Do Dora report a decade ahead. Dora also predicts as a technology landscape continues to evolve.

Con DevOps community is committed to do the fundamental shift and the fundamental principles that have always been part of the DevOps movement, which is culture, collaboration, automation, learning, and using technology to achieve business goal, what stake can stay around? So what they also have reflected is, uh, when the DevOps label exists and does it matter. And of course, I mean, if you see that during a decades of, you know, fostering that collaboration and you know, bringing DevOps forward with, you know, uh, the mindset of the practitioners and the leaders, we have seen that it has become a status scope.

So now what is next for DevOps practitioners and, and organizations who are looking at like fostering more collaboration through DevOps practitioners? I would say DevOps plus for engineering leaders, and I will indicate four definite, you know, areas where DevOps leaders or engineering leaders should pay attention. The first is the big leap of open source.

I will talk more about it. I am engaged in, uh, various open source initiatives. So I have a very detailed outlook of, uh, the open source community.

We also see a lot of traction with data, machine learning and ai, and how DevOps practitioners will not only leverage data machine learning and ai, but also foster more collaboration with AI to deliver better outcomes, security and regulations with, uh, security and regulations. A lot has been changing in the past year and people have started to talk about post quantum security and regulatory, um, you know, uh, actions. And also, uh, a lot of, uh, more tooling needs to be kind, kind of in be in place.

So we can talk about that, uh, in detail. And the fourth pillar is DevOps tools and technologies. So what is the next, you know, next futuristic things, which what, you know, craft, uh, you know, the next five years for DevOps practitioners?

That is also something which we will talk about during this talk. So projecting the possible future with open source, open source solutions, uh, are becoming more and more popular with businesses and organizations as they look towards cutting cost, reducing vendor lock-in and increasing flexibility. There has been a noticeable trend in some of the areas, for example, licenses.

So permissive licenses for open source, uh, is, uh, uh, trending in contrast to copy left licenses, particularly those in the GPL family, which have decreased in usage permissive licenses, imposed minimal restrictions on software users, allowing them to incorporate the software into proprietary applications without disclosing the source code. So this is one of the key areas of, you know, um, uh, futuristic areas to look at. Licensing models have, uh, um, you know, there will be a lot of, uh, changes and a lot of, uh, you know, a new, um, you know, uh, action on the licensing from the licensing community.

So we have to watch out for that. Open source software represents the paradigm shift that fosters collaboration and transparency and community driven innovation. 9% from 2032, uh, 2023 to 2030.

According to a new report by Grandview Research, DevOps teams often use open source tooling and platform to streamline development workflows, improve efficiency, and leading to increased demand for open source, uh, uh, services. And if you are not hiring beneath the rock, you must be following the open source movement, uh, which is happening in the DevOps, uh, ecosystem. There are many big, you know, and small, uh, companies, applications, tools, which are fostering that, uh, collaboration for the DevOps, uh, organizations.

We will also talk about some key trends, uh, on the, uh, in the open source, uh, ecosystem, starting from localized open source. And, you know, this is, um, misunderstood term. So when we talk about localized open source, the open source software committee is witnessing a remarkable surge in adoption across developing countries.

And that is what this whole, you know, paradigm shift is reflecting. This trend is driven by the need for cost effective, scalable solutions that can be tailored to locally. With 80% of companies reporting increased utilization of open source technologies over past years, it empowers local developers to contribute to global projects, enriching the community with diverse perspectives and innovation.

Another trend is AI driven open source projects. And, uh, if you are con uh, contributing to open source projects, you should pay attention to this. The integration of AI and ML into open source software is a significant trend and primarily driven by the need of more efficient and intelligent development processes.

The AI integration of AI ML in open source tooling is not just a trend, but also transformation offering at times either completely new capabilities or new ways to contribute. So watch out for this trend. Cross industry support and sponsorship.

So obviously, um, open source is, uh, uh, being adopted by many industries as we speak. Many industry segments cross industry partnerships have become a hallmark for open source software communities. Companies from diverse sectors are joining forces to leverage open source technology.

And lastly, I would say community to commercialization. So business model for open source is also maturing from support and services to open core in the past. And now commercialization of open source.

We'll see developer driven growth instead of relying on founder roadmap. And in in turn, what it will mean is new licensing models will emerge and we will see a shift in the traditional approaches of, you know, how open source licenses, uh, were working and what kind of new models emerge from there. Developer driven evolution of more commercial models for open source.

Also, uh, coordinating a lot of organic growth of open source projects and leading to enterprise sales will. Uh, and it's also driving new business models and new roles in the open source community. There's an article, uh, which is, um, you know, uh, at the bottom of the slide if you're interested in, uh, this topic more you can refer to this article.

So let's, uh, talk about now more opportunities and roles into open source, um, ecosystem. So one of the opportunities which I see is expansion of open source program offices. And, and this will bring new job opportunities.

For example, if you have not heard about o spoke plus, plus it's institutionalizing open source globally. Uh, OPO plus plus is a global network and a community of collaborative open source program offices focused on supporting the core mission of an objective where universities, government, civic organizations come together and, uh, join forces and collaborate, right? So there is also a link in the slide if you're interested to know more about it, open source implementers and advisors.

This is also a new segment of job roles, which is in creation or in making to support practical action oriented approach that helps address the nuances and challenges of open source, uh, technology adoption, let's say, within government, within public organizations, with within more restrictive, you know, mission critical, um, landscape and, uh, regulated sectors. So you will see a lot of these job roles coming up and in demand. Another interesting area is open source, uh, support desk, a new segment of professionals which will ensure a smooth user experience and enhanced collaboration with open source project communities.

And lastly, I would say open source academy, building open source skills and awareness and set up a community of practice, for example, for an organization or a public sector or policy makers. It's also on the card. So these are very interesting areas, um, to look into.

Now we also have talked about the second pillar, which is a data machine learning and AI and big game changer AI is likely to be. 7 trillion to global economy in 2030. So what it means for develop DevOps professionals and how should we, um, you know, assess and also leverage this movement.

So I would suggest that, you know, start looking at DevOps and AI as you know, two, uh, companions in this journey. DevOps, a collective journey towards evolution of software practices and AI enhance predictive and adaptive decision making. So new ways to use data sources, for example, how do you inculcate, uh, that kind of philosophy into your, uh, day-to-day DevOps, uh, you know, practices, the approach to human machine interaction.

How do you collaborate with, uh, the open source development community, the 19 million plus developers out there? Um, we have talked about open source and also, um, uh, we also see that it's not only faster, but more valuable software, which counts, right? So all this has also, um, um, made us think as leaders that how should we design our organizations, which are AI driven?

So you will see a lot of AI native companies, you know, spinning off from this AI movement. So designing an AI driven software engineering team, how could it look like? Um, there is a lot of initial investment in developing training and integrating AI systems for software development.

And that can be substantial. Not only, uh, initial investment, but also continuous maintenance update training of AI systems to keep them effective can also incur significant cost. So what we will see, um, in the near future is that, um, obviously ai, assistant coding is on the cards for every organization.

We also have seen a lot of traction and, uh, you know, trends around agents, right? So how software development agents or product management agents or operation support agents or research agents would help us crafting or designing an operational model for an AI driven software engineering team. It'll be very interesting for the leaders to kind of take a deep dive into, uh, the, uh, current of existing operational model and see how it can shift from a technology capability perspective, how we can bring more, uh, technology, uh, companions into our journey for software development and lifecycle management.

And AI agents or AI system code coding would also be part of your operation model. So obviously new methods are brought prospected to be considered in the software development processes. Um, it also will, um, create some new possibilities for easy and human friendly software obstruction layers, which could make the coding process much more easier.

And it can be a job for any person. So AI systems and moderators, we have talked about that, um, will be one of the new job roles, which, uh, can kind of, you know, help your organization steer the needle in the right direction and foster that collaboration for the next generation operating model. AI systems also are envisioned to, uh, to be used for creating codes by exporting or reusing already existing codes.

So obviously that movement had already started, but we will also see a shift in that approach, how the legacy code can be transformed into, you know, next generation application. That is something which we will also see, uh, happening as we go along with this movement. And lastly, I would say hyper assistance, uh, suggestions for improving developer working routine by 2030.

So, um, things would obviously substantially shift or change and upscaling would be needed in order to sustain that technology revolution, which we are seeing. So as leaders, what our job would be is to balance the effort where into securing AI initiatives, while building momentum into AI based innovation, collaborating with security experts early and often in the development process. How do you protect, uh, US against AI vulnerabilities, for example, specific threats for AI environments like modeling, invasion or data poisoning, or functional extraction and advancements like malicious cousins of charge GPT and, uh, for example, bomb GPT, right?

If you haven't heard about it, you can kind of search, uh, this, these steps. Integrating evolving regulatory requirements, ensuring responsible use of third party applications and mandating partners to use responsible AI and creating an evolving AI security processes and policies aligned with existing upcoming regulations. So for, from a practitioner perspective, uh, there is a lot happening from a leader's perspective also, it is a lot on the table.

Now we also talk about technologies and their impact on job creation. So there was a report by World Economic Forum, a 30 to 35% increase in demand for roles such as data analysts, scientists, big data specialists, business analysts, data and network professionals, and data injuries that is driven by the advancement and the growth in the adoption of frontier technologies, which rely on big data. So I think that would also foster some change in terms of how DevOps professionals take that next leap or next step into the future.

Now we talk about the third pillar, which is evolution of practices and what DevOps plus would mean. And I have, uh, on the right hand side put the hype cycle of Agile and DevOps, and I have also looked at site reliability engineering, uh, hype cycle 4 20 24. And after researching, I've put some of the interesting concepts, which I think I'm very excited about.

Uh, 4 20 25. Um, I've divided this into two parts, operational practices including, uh, practices like and tools like open SLOs or SLO management, AI assistance, law monitoring and resilience or automation. And the second part or second, uh, aspect is feedback loop.

So how do we, um, optimize delivery from by managing progressive delivery, observability driven development, managing AI generated trust, et cetera. So these are, uh, some of the interesting aspects, which I will talk about today. So some developments, uh, for DevOps practitioners in terms of tooling applications, what kind of changes you would see in 25 infrastructure from code, not infrastructure as code.

So what we have seen is infrastructure as code is where you as a developer needs to explicitly define infrastructure resources in a separate file, say cloud formation server as framework YML files or CDH stats, right? Infrastructure from code is a way of creating applications such that at deploy time, your cloud provider inspects your application code and then automatically takes care of your provisioning, whatever underlying infrastructure your application code needs. And this is not science fiction.

There are tools available already in the market, uh, for such kind of self-provisioning capability for infrastructure from code. So new tools like anchor, shuttle, mortal, all these tools, um, um, are available. And of course there are references and there are materials to look at.

Um, I have, uh, some links in the slide if you are interested to, uh, read through. I also talk about policy code. Policy code is not new.

Policy code refers to a practice of managing and implementing policy decisions through code and making that enforceable and verifiable in the environment. Why we are talking about this today is, uh, due to the changes in the possible regulatory, the security posture, and also aligning a lot of like process around, you know, practitioners, it would be more relevant and it becomes a lot more critical that we start to rely on common set of as code skills for policy. And, um, we will see more and more of this, uh, in 2025 and beyond.

I talk about, uh, some key considerations around cost. Cost are at crossroads for organizations who have started to spend, um, on cloud. So currently cloud spending represents approximately 30% of overload.

3 trillion by 2025 according to Dave McCarthy. So unmonitored cloud usage, for example, our unused cloud licenses, which are left, uh, accidentally during night times or weekends, or variable cycle costs, depending on the number of processes used to run code, et cetera. Um, these will add up to your cost.

And that, um, is something which, uh, you know, the leadership, uh, people should watch out for. Gartner also advises CIOs to negotiate commercial terms with their cloud providers, such as volume or time-based commitments and other options to move from on demand to resolve like instances. There are other things which are happening in this space, which I will not talk about, uh, today, but there is a blog which I have written for, uh, the strategic shift from cloud first to cloud minimalism, and now post minimalism, like how people are reacting to this AI movement and what is entailed for us as leaders when we talk about capabilities, um, of data and, you know, in, uh, leveraging AI into our, you know, ecosystem and how we foster that collaboration with cloud providers.

Augmented finops is another area where, you know, there will be a lot of action, which is building upon traditional finops by infusing it with artificial intelligence and machine learning. And these technologies enable autonomous and, uh, continuous optimization of data infrastructure shifting for, from reactive cost management to proactive strategic planning, AI assistance. We have talked about it in the past, uh, slides.

So, um, there will be a lot of traction around artificial intelligence infrastructure as code, a new approach to automate, uh, the generation of, uh, let's say infrastructure as code templates, configurations, utilities, uh, queries and more using open AI APIs, for example. So there will be a lot of, uh, you know, traction, uh, on this through AI assistance. I'll talk about open observability.

One of my favorite topics for 2025 open observability is the idea of capturing telemetry signals from many different technology, leveraging open standards, open source tools, application program interfaces. And a lot is happening in this space. 70% of new cloud native applications, uh, will adopt open telemetry for observability rather than vendor specifications.

So open observability becomes, uh, even more important. There is, uh, other ideas like bring your own storage backend to enable more flexibility and efficient utilization of infrastructure insertion of LMS to simplify the user experience and enhance both analytics and downstream automation. So there is a lot of action happening in this space, one of my favorite areas for 2045.

Now, we are coming to the end of, uh, this discussion. There's a lot more to talk about, but I will conclude my talk, uh, with this slide. Like, how will technology jobs look like in 2030?

So, uh, I'm quoting McKenzie and co cooperation titled Skill Shift Automation and the Future of Workforce, uh, demand for Higher Cognitive Skills such as Creativity, critical thinking, decision making, and Complex Information Processing will grow through 2030. Some of the practical tips, if employees who are journalists with broad skills, with no contemporary areas like cloud or cyber, et cetera, of subject matter expertise, will likely struggle to remain relevant in current and future positions. Those who excel in individual programming languages currently, uh, might feel that sense of security, but there's a need to understand multiple languages and it'll grow quickly in the next five years.

And some of the IT skills which are nearing end of life cycle, including let's say manual testing or you know, a square and those kind of things, I think these are being progressively replaced by ux, ui, cross-functional team members with automated test skills and cloud-based engineers. So probably things to watch out for concluding my talk leadership in the age of AI is critical leaders play an important instrumental role in creating effective governance strategies built upon various pillars. As discussed in this talk, DevOps still remains a center of focal point or a core fundamental capability to enable that ecosystem of emerging technology.

DevOps practitioners should look out for the facelift. Try to include, uh, a lot of progressive tools in your, um, capability. Um, also try to kind of, uh, see what kind of upskilling meets you, you and your staff has, and, uh, start to think about the new operating model with these AI assisted tools, AI agents, and how do you foster human machine collaboration.

With that, uh, we come to the end of the talk. Thank you for listening today, and have a great day at 2025.