National Coding Week is a timely reminder that, even as AI reshapes software development, strong coding skills remain fundamental to innovation. AI-assisted tools like GitHub Copilot, Cursor and Claude Code are now producing code faster than engineers can review it, precisely why deep technical judgement matters more, not less. AI can generate and accelerate, but it can’t be accountable. Engineers still need to understand what these tools produce, catch the code that’s ‘almost right, but not quite’, and take responsibility for what ships.
As AI systems become more capable and autonomous, the skill that differentiates engineers is no longer whether they can write code but whether they can design, verify and govern AI-assisted development. We’re moving from engineers who are ‘in the loop’ at every step to engineers who are ‘on the loop’. They’re defining what ‘done’ looks like before code is generated, setting the guardrails, and stepping in when quality, security or accountability is at stake. That raises the bar on technical expertise, which is why effective skills management is now critical. Organisations need visibility into the coding, AI and critical-thinking skills across their workforce so they can close gaps, strengthen oversight, and keep humans accountable for the outcomes AI produces.
