What we observe about hiring once candidates work with AI — findings, changes, and insights. No spin, just what we learn. It's early, so this is a short list that will grow as real assessments run.
When every application is written by the same handful of AI tools, the document tells you about the tool, not the person. The thing worth measuring moved from what someone wrote down to how they actually work a problem — which is exactly what an interactive challenge can show.
In early runs the transcript carries most of the signal: which questions they ask first, what they verify before trusting the model, and when they decide they're done. Two candidates can reach the same answer and read completely differently — and the difference is the part worth hiring for.
A fixed question bank ages badly and leaks. We moved to scenarios and rubrics authored per campaign from a short brief — so the challenge matches the actual role, the weights reflect what that job needs, and every candidate for it works the same problem under the same rules.
More as the experiment runs. Curious how the scoring works? Read the framework →