Your library. /admin/assessments lists the assessments your workspace owns. Create one, edit the draft, publish it, and archive it when you're done.
Two ways to make one. Clone a ready-made template for a role at one of three depths — *screen* (5–10 minutes), *standard* (20–30) or *deep* (45–60+) — or answer a short intake about the role and have a tailored set generated (this uses AI actions), sized to the time you allow.
Question kinds. Short answer and multiple choice (about a minute each), long answer (about three) and code (about five).
Sending. Invite a candidate to a published assessment and they receive an e-mail with their own link to take it.
Grading. Multiple-choice and short-answer questions grade themselves. Long answers are read by a person. Code runs in the candidate's browser against your test cases and reports how many passed; the platform checks the count is plausible but does not run the code itself.
Candidate self-assessments. Separately, candidates can take four optional questionnaires — a Myers-Briggs-style type indicator (20 questions), Leadership (30), Work Style (15) and Communication (15). Together they add at most 25 points to the application-strength score.
Posting and reviewing are free. Every stage, e-mail and export described here is in the console from the first sign-in.