AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Why it matters
Users routinely add requirements only after seeing a coding AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition →'s first attempt. This quantifies the cost: late-arriving requirements cause roughly double the code rework of ordinary edits, a concrete reason to front-load requirement elicitation in agent workflows.
articleA Few Pages of Markdown: Committed AI Configuration and Lower Quality Cost after Coding-Agent AdoptionYegor Denisov-Blanch, Shyam Agarwal, Pavel Azaletskiy, Hao He, Rylan Schaeffer, Brando Miranda, Bogdan Vasilescu, Sanmi Koyejo
articleEngineering Reliable Coding Agents: Evaluating and Operating the System Around the ModelStephanie Jarmak
articleSpecification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code reviewJoel Abenhaim