AI coding tools reshape software engineering but some developers miss the challenge
CNBC reports that AI coding tools are speeding up development, yet many engineers feel the work has lost its sense of struggle and worry about the impact on junior talent.

Key takeaways
- AI coding tools are now used by the vast majority of developers, dramatically speeding up prototype creation.
- Many experienced engineers report a loss of the deep, hands‑on problem solving that gave their work satisfaction.
- There is concern that junior engineers will miss crucial learning experiences if routine coding is fully automated.
- Industry leaders stress the growing importance of judgment skills and the need to embed AI into training rather than replace it.
In this article
AI tools accelerate development
According to CNBC, AI coding tools have become a daily part of work for the majority of software engineers, with a 2026 Stack Overflow survey showing 83 per cent of respondents using at least one such tool. Engineers can now generate multiple prototypes in an afternoon instead of spending weeks on a single version, dramatically increasing output speed. However, the volume of generated code means engineers often cannot review every line, shifting the core skill from writing code to judging which suggestions are safe and useful.
Developers miss the hands‑on challenge

Long‑time developers told CNBC that the satisfaction of tracking down a stubborn bug over days has been replaced by a quick prompt that yields a solution in minutes. Gemma Rodgers, a UK‑based developer, described the feeling as a "breadcrumb of joy" rather than a full meal, noting that the flow state of deep coding sessions is harder to achieve. Others echoed this sentiment on social media, saying the creative struggle that once defined the job is fading.
Junior engineers risk losing learning opportunities
Routine coding and debugging have traditionally been the training ground where junior engineers build experience and develop problem‑solving instincts. Chris Sees, a veteran engineer and CEO, warned that if AI takes over these tasks, newcomers may miss the "scar tissue" that comes from fixing real errors. Rodgers added that core investigative skills could erode if engineers rely solely on automated suggestions, especially when tools or services become unavailable.
Leaders debate productivity versus skill retention
James Broadhead, CTO of Taktile, highlighted that the new bottleneck is judgment — deciding which AI‑generated prototypes are worth keeping. Joshua Wöhle of Mindstone observed a widening productivity gap, describing the emergence of "100X engineers" who master AI assistance versus those who struggle to adapt. While some leaders see AI as a capable junior assistant that frees senior staff for higher‑level problems, they agree that companies must integrate AI into learning pathways rather than allow it to become a shortcut that bypasses essential skill development.
Sources
Spotted a mistake? See our corrections policy. Read how News by OnDevice works in our editorial policy.


