Align four agents' output
Your team of four is building out a new Java microservice, and each engineer uses an AI agent for their slice. After a week the service is a stylistic patchwork: one agent used constructor injection, another field injection; one used checked exceptions, another runtime; error responses have three different JSON shapes. The code works but reviews are painful and the API is inconsistent. How do you get four AI-assisted engineers producing coherent, consistent code?
find_style_drift(module_lines: list[str], exemplar_line: str) → list[str][["orders|constructor|runtime|api_error","billing|field|checked|billing_error","search|constructor|runtime|api_error"],"service-template|constructor|runtime|api_error"]out["billing|injection","billing|exceptions","billing|error_shape"][["orders|constructor|runtime|api_error","billing|constructor|runtime|api_error"],"service-template|constructor|runtime|api_error"]out[]Treat the AI’s output as a draft to verify, not an answer to trust. Name the specific flaw and the input that triggers it, say how you’d catch it (tests, edge cases, reading critically), and how you’d re-prompt or decompose to get it right.
Vibe & agentic: describe the solution in plain language (or narrate it) and the coach grades your approach.