A code completion tool suggests the next few lines while a developer types. A coding agent takes a task, such as "add input validation to this endpoint and update the tests", and works through it: reading files, making changes, running commands and checking the results, often over many model calls. That makes it more capable and more expensive per task.
Because the cost now follows usage, engineering and finance leaders need to read spend per engineer as a distribution, count the review and rework agent-written changes create, and measure value at the level of merged changes that clear a quality bar. Managing Coding-Agent Spend sets out the measures and guardrails, and What the Engineering Evidence Says About AI Coding Value summarises the research.
