DevOps professionals will be under pressure to learn more new skills than ever before – as context engineering touted by Gartner as the next vital skill
Gartner has identified context engineering as critical for AI-enabled processes. This step beyond prompt engineering aims to improve AI outcomes and reduce hallucinations. For instance, when using an LLM in deep research mode, context engineering can include prompts, access to supporting documents, systems, and other relevant information, as well as instructions on its role, such as that of an analyst or researcher. However, this represents another skill that teams will need to acquire and so DevOps leaders must prioritize giving them this assistance.
Vibe coding becomes much more than just coding and turns software engineers into DevOps superhumans responsible for everything from requirements through to deployment
Vibe coding will extend beyond more than writing code. From within their IDE, a single DevOps team member will be able to AI to supervise everything: from supervising to development, testing (‘vibe testing’), and deployment. Traditional steps in the software development process will disappear, roles will change and blur BUT software engineering principles and being a good software engineer won’t change. This is essential and often overlooked.
But, critically, vibe coding will depend on teams still having a strong understanding of software engineering principles (so that they understand what AI is doing and avoid errors), alongside acquiring new skills like context engineering. So, it’s time to sharpen up on those skills if not already in place.
Say goodbye to the need to learn programing languages, as DevOps roles will shift up more than people expect
Shift-up means three things to me.
First, with autonomous tools, many of the traditional phases in the DevOps process will disappear, as they are no longer needed with the advent of AI (it’s already happening with writing test scripts – they are now not needed).
Second, again, with AI, DevOps professionals will no longer need to learn programming languages, and instead, will use natural language to supervise processes, with one person able to cover everything, across requirements, planning, development, testing, and deployment. So that also means no more hand-offs between teams and the blurring of traditional DevOps roles.
Third, with all that in place, every engineer will be able to shift up and self-promote themselves into more senior roles, giving them career progression and potentially new opportunities within their organizations. Now, supervising a team of AI agents, any engineer – developer, tester, or SRE – will be able to act as an architect or very senior principal engineer.
However, for this shift-up to be successful in practice, it will be vital for them to have knowledge of software engineering principles. Without understanding what AI is being asked to do, humans cannot evaluate the quality of the output, which in turn can lead to risks such as vulnerabilities that could result in security breaches. In 2026, humans remain as important as ever in software development, but only if they understand what they are asking AI to do and AI’s output.
Quantum attire will revolutionize the retail industry
AI will transform different markets in previously unexpected ways. For instance, Gartner has spoken about quantum attire, the concept of having one outfit that AI can modify according to factors like mood, weather, event, and more. It could mean the end of fast fashion and impact retail employment, as well as associated areas such as media, advertising, and pop culture. Malls could continue to decline when there is no need to go shopping. Attire will be a form of communication that is tech-dependent. Of course, software-dependent innovations like this will require rigorous DevOps processes, particularly the strategic and carefully controlled use of AI, especially guardrails around data privacy and security.
AI gives healthcare the upgrade it deserves
Through AI innovations, we will be able to conduct trials and perform drug discovery in increasingly shorter timeframes. It’s already happening: compared to the typical 2.5-4 years in traditional drug discovery, Insilico Medicine reported earlier this year that it has used generative AI to bring the process of developing drug discovery down to 12-18 months on average, from project initiation to nomination of preclinical candidates.
With AI, doctors will finally be able to reduce the massive paperwork burden that takes up so much of their time. At the moment, a 30 minute patient appointment might require 20 minutes of documentation for insurance and compliance purposes, but generative AI-powered documentation will automate clinical note creation, feeding it into the right systems, and freeing up doctors’ time for more patient care. Morever, the richer, more consistent documentation generated by these tools will enhance continuity of healthcare, as patient records move across physicians’ and other healthcare systems, helping to repair what has become a highly disjointed approach.
