From theory to practice: How organisations are adopting AI
Following up on my recent post about the real impact of AI on engineering leadership, I came across an interesting article from the Stack Overflow blog. Titled “Not an option, but a necessity”, it moves beyond the theoretical and shows how different companies are practically integrating AI into their workflows.
The article reinforces that while many leaders see AI deployment as immature, the push for adoption is undeniable. It is driven by the need for greater efficiency, faster access to knowledge, and more innovative strategies.
How different companies are approaching AI
What I found most useful were the specific examples of how different organisations are tackling this challenge. Each has a distinct, intentional approach.
Cloudflare: They are taking a top-down approach by giving AI tools to their most senior, high-performing developers first. The goal is not to make average developers good, but to make the best developers even better. This allows them to measure the productivity impact on complex problems before rolling it out more widely.
GitHub: Their focus is on freeing up developer time for more strategic work and collaboration. By automating routine tasks, AI allows their technical teams to focus on higher-value activities and work more closely with non-technical departments, breaking down traditional silos.
Abnormal AI: They take a pragmatic view, acknowledging that AI tools do not work perfectly “out of the box”. Their approach emphasises that success requires investment and a willingness to shape the tools and the organisation to fit each other.
A systems engineering perspective
From a systems engineering perspective, these approaches are fascinating. Instead of treating AI as a simple plug-and-play tool, these companies are managing it as a new, powerful component within their complex organisational systems. They are carefully observing the interactions—the feedback loops—between the technology, their people, and their processes. This allows them to manage the emergent outcomes, rather than just hoping for the best.
The common thread is strategy
These examples perfectly illustrate the point from my previous post: successful AI adoption requires a clear roadmap. Each of these companies has a deliberate plan. They are not just adopting technology for its own sake; they are integrating it to solve specific business problems, whether that is amplifying top talent, fostering collaboration, or accepting the need for customisation.
It is encouraging to see these practical approaches emerge. They confirm that the path to getting real value from AI is less about the tool itself and more about the strategy behind its implementation.
