Maturity varies widely among vendors claiming to offer engineering AI agents. Engineering leaders should ask five key questions: Is the agent workflow-specific or vaguely general? What data does it actually access — CAD, drawings, standards, or just text? How are system-level instructions and organizational context provided to the agent? Has the agent been tested and refined using real engineering data? And how does the vendor ensure predictable, reliable outputs over time? The most mature agents typically come from vendors with deep experience in the specific workflow itself — for example, design review agents built by companies that have supported design reviews for years, or simulation agents developed by simulation software providers.