The first answer is ready. The follow-up is not.

Career-readiness problem

Placement interviews punish the polished retell.

A learner can rehearse the opening. The live moment is the second question—the follow-up that asks why, how you know, or what you would change.

Socratic AI Interviews require an explanation, then a pushback. Interview Sequences put several of those interviews in order when the process has stages. Shared Footprint is how they share selected attempts, including gaps, with a coach or career team.

Career Services Teams run this as a cohort. Career Coaches inspect one client’s record. This page is the problem: interview practice before placement. The record is practice evidence. It is not certified competence, and it is not a hiring score.

Fit check

Who interview practice is for — and who should walk away.

If your only success metric is “students finished the mock module,” stop here. Completion is cheap. Retrieval under follow-up is not.

• Career Services & placement

  Directors who need interview follow-up evidence before the next drive—not another completion report or scripted quiz score.

• Career coaches & advisers

  People who open Shared Footprints between appointments and coach from attempt → feedback → named gap, not from a polished paste.

• Wrong buyer jobs

  Employer screening stamps, LMS replacement, credential wallets as the primary deliverable, placement-rate promises, or competence certificates sold as hire/no-hire.

What you get

A bounded interview pilot that leaves inspectable evidence.

• One role × one date × one reviewer

  Name the interview the cohort still fails in the room. Access goes live for that sitting—not a campus-wide catalogue walkthrough or Gem-first path.

• Socratic AI Interviews (or a Sequence)

  Learners answer first under stated conditions. Add Interview Sequences only when the live process has more than one stage.

• Shared Footprints with gaps

  Selected attempts, follow-ups, revisions, and what is still missing—accepted, missing, or not assessed—for human review.

• Keep / change / stop

  A readout at the end. No pressure to expand seats before you have seen the sittings and judged reviewer load.

The loop

Explain, take the follow-up, share the gap — then decide.

• Days 1–3 · Name the interview

  Lock one role, one target date, and one reviewer who will open the attempt. Map an optional Curriculum Graph goal so practice attaches to a capability, not a random prompt dump. Output: a one-page scope sheet for that sitting only.

• Days 4–10 · Sit the practice

  Learners sit a Socratic AI Interview—or an Interview Sequence if the live process has stages. Require a short prep note before the first sitting. Look for explanation under follow-up, not a rehearsed opener pasted in.

• Days 11–16 · Inspect the follow-up

  Your reviewer opens a sample of Shared Footprints between appointments. Read what changed when the question moved. Name the gap in the learner’s words. Ask: would you coach from this trail?

• Days 17–21 · Share and decide

  Shared Footprint can include gaps; the reviewer still decides what it means. Three questions only: Did footprints show behaviour your current artefacts miss? Was load tolerable? Keep, retune the prompt bank, or stop.

Capabilities in this loop

Two interview products and a shareable record.

• Socratic AI Interviews

  A claim, then the follow-up. Not a scripted quiz. Learner answers first under stated conditions.

• Interview Sequences

  Several interviews in a planned order when the live process has more than one stage.

• Shared Footprint

  Selected attempts, including gaps. Practice evidence only—not verified competence.

Boundary

The follow-up, not only the opening.

• Attempt trail

  What they said first, and what they said when the question changed—enough signal to coach without inventing a brochure percentage.

• Reviewer

  A coach or career team who inspects the record between appointments. They still coach; ProfilEd does not score candidates for employers.

• Stack fit

  Many pilots start with named accounts—no campus-wide SSO required to see whether sittings work. Bring IT when you need SAML/OIDC or LTI.

• Practice before placement. Not a hiring product.

  ProfilEd records interview practice. It does not certify competence, match people to jobs, predict placement %, or treat the record as recruiter-auditable proof. Humans interpret Shared Footprint—including gaps.

Evidence rule

What counts in this pilot — and what does not.

• Counts

  Stated conditions; attempt → feedback → revision; gaps marked honestly; human interpretation of the Shared Footprint; reviewer load you can sustain between appointments.

• Does not count

  Completion %, time-on-platform, or “AI confidence” as competence; polished answers pasted in; claims that two weeks verify programme-wide readiness or raise placement %.

• Out of scope

  LMS/SIS replacement, campus-wide rollout on day one, employer screening as the product, credential wallets as the primary deliverable, invented board-slide percentages.

• Decision language

  Write one sentence at readout: expand next cohort; retune mode/sample rule; or stop. Stopping is success if it prevented a bad annual licence.

Continue

See the interview product, who runs a cohort, and the method.

• Focused pilot

  Primary CTA—bounded evaluation.

• Socratic AI Interviews

  Explanation under follow-up.

• Watch AI Masterclass

  Observe a session before sitting one. Public; no account required.

• Career Services Teams

  Cohort workflow around this problem.

• Interview-prep cohorts

  Add Study Rooms when you need a join-by-code slot.

• ProfilEd methodology

  Practice, challenge, revision, and human review.

Next step

Bring one interview date and one reviewer.

Plan a focused pilot