As AI reshapes software engineering, the role of developers is shifting from writing code to defining requirements, validating outputs, managing risk, and ensuring quality. Thoughtworks' report argues that engineering rigor doesn't disappear with AI—it moves upstream into specifications, testing, governance, and security. Traditional practices like code reviews are evolving into specialized functions focused on mentorship, consistency, correctness, and trust, while a new "middle loop" of supervisory engineering emerges to oversee AI-generated work. The report also highlights the need for stronger agent security, modern technical foundations, and continuous experimentation, emphasizing that organizations must redesign engineering practices, roles, and operating models to fully realize AI's potential without sacrificing software quality or resilience.

Join for free to read