RecruitLingo

Here is a take-home assessment we send candidates for a {{Job Title}} role. THE ASSESSMENT: {{paste the full instructions, plus a description of any starter code or data}} WHAT IT'S SUPPOSED TO TELL US: {{the skills or judgment calls it was designed to test}} OUR POLICY ON AI USE DURING IT: {{allowed / allowed if disclosed / not allowed / never stated}} Attempt it the way a candidate using an AI assistant would. Then tell me: 1. How close your output gets to a strong human submission, and what a reviewer would have to notice to tell them apart. 2. For each part of the exercise: trivial for AI, partly assisted, or still dependent on judgment and context you didn't have. 3. Whether, given point 2, the exercise still measures what it's supposed to. Be blunt if it doesn't. 4. Two or three changes that make it harder to fully outsource without making it longer — a decision with a trade-off to defend, messy or contradictory inputs, a short written rationale. 5. A 15-minute follow-up conversation plan that would expose a submission the candidate can't explain. 6. One line I could add to the instructions that states our AI policy plainly.