A leading software productivity expert, Dax Raad, has warned that while artificial intelligence (AI) is rapidly automating coding tasks, a new productivity bottleneck is emerging around work coordination and decision-making rather than the act of writing code itself. In an interview and commentary shared in Business Insider, Raad argues that traditional measures of developer productivity — such as lines of code written or features delivered — are becoming less relevant as AI tools take on much of the mechanical work. Instead, the bottleneck is shifting to human roles that involve planning, prioritising, managing complexity, and integrating AI outputs into larger systems.
According to Raad, AI’s ability to churn out functional code quickly has reduced the time spent on routine programming chores, which historically consumed the bulk of many developers’ days. This shift, he says, means organisations must now rethink how they assess productivity and structure work: rather than rewarding individual coding output, companies will increasingly value skills like architectural design, cross-team coordination, clear communication, and strategic problem-solving. In this new paradigm, developers may spend more time orchestrating AI systems and aligning technical work with business goals.
Raad also notes that as AI accelerates code generation, decision-making processes and collaboration overhead can slow teams down if they are not properly organised. For example, teams may find it challenging to decide which problems to have AI solve first, how to structure generated outputs, or how to reconcile different AI suggestions — tasks that require human judgment and consensus. This new set of constraints could replace coding as the primary drag on productivity, particularly in large or distributed teams where alignment and prioritisation are already difficult.
Ultimately, Raad suggests that to thrive in a world where AI handles much of the mechanical work, companies need to invest in tools, training, and organisational processes that support better coordination, clearer decision-making, and effective human–AI collaboration. Rather than competing with AI in writing code, humans may increasingly act as designers of workflows and stewards of complex projects, ensuring AI outputs align with broader business strategy and user needs.
