Optional Capstone · ~45 min · Advanced
The Full Interview — End to End
One candidate, one complete interview, start to finish: review the role, ask behavioral and situational questions, probe for evidence, take notes, complete the full scorecard, run the case station, and make a defensible recommendation. This is the whole job in one sitting.
AI Instructor: "Optional, but this is the closest thing to the real chair. Treat the AI candidate like a real hire — the habits you run here are the habits you'll run on a Tuesday at 2pm."
How the capstone works
- One continuous flow — seven stages, no teaching breaks.
- You assess all eight competencies across the interview + case station.
- Five agents run it: Interview Simulation, Candidate Role-Play, Case Simulation (#8), Scoring/Calibration, Feedback Coach; the Final Readiness Evaluator composites the result.
- Scale-based. You get a readiness band + targeted development guidance — never a pass/fail.
- Optional: skipping it does not affect certification; completing it can surface Expert-band evidence (calibration leadership).
Stage 1 — Review the role context
The Interview Simulation Agent briefs you like a real prep packet:
- Role & discipline: MD/DO, NP, or PA (sets the scope/autonomy calibration).
- Site context: UC clinic profile — volume, solo-coverage hours, team makeup, after-hours reality.
- What's already known: TA phone-screen notes (licensure, interest, logistics) — so you don't re-screen.
- Your charge: assess Competencies #2–#8 in the interview; #1 in the case station. Plan your questions.
"Plan before you talk. Which competency does each question target? Where will you probe?"
Stage 2 — Behavioral questions
Open with behavioral questions from the approved bank ("Tell me about a time…"). The AI candidate holds a hidden persona + proficiency and won't hand you clean STAR answers.
- Cover the behavioral-leaning competencies: #3 Professionalism, #4 Teamwork, #5 Coachability, #8 Values (and #2 Communication as it surfaces).
- Listen for STAR/SOAR completeness; note the "I vs. we" tell.
Stage 3 — Situational questions
Move to situational/SJT items for the UC-judgment competencies the candidate may not have a past story for:
- #6 Autonomy (9pm solo transfer-borderline), #7 Adaptability (three high-acuity arrivals, one nurse), and a #2 Communication de-escalation scenario.
- Cross-check against the behavioral answers — does what they'd do match what they've done?
Stage 4 — Probe for evidence & capture notes
Throughout Stages 2–3, the simulation enforces the core discipline:
- Probe every thin answer — depth → evidence → reasoning → contrary. Silence is allowed.
- Capture evidence notes (quotes, not verdicts) in the note panel; the agent nudges if you start scoring on vibe or writing inferences.
"If you didn't probe it, you can't score it. I'll flag any answer you accepted at the surface."
Stage 5 — Complete the scorecard (all eight competencies)
Score each competency you assessed on the BARS 1–5, citing evidence from your notes. (Competency #1 is scored next, in the case station.) The AI holds your scores until the case station is done, then reveals the full calibration read together.
Stage 6 — Clinical case / chart-talk station (Competency #1)
The Case Simulation Agent (#8) presents a role-calibrated common UC case. You run it as the clinician rater:
- Present the case; ask for a think-aloud; reveal info only on request.
- When they commit, the agent drops one wrinkle — watch for re-reasoning vs. anchoring.
- Probe reasoning, not recall; score Competency #1 on reasoning & safety, role-calibrated (consultation = green flag for APP scope).
Stage 7 — AI scoring, calibration feedback & final recommendation
Now the full read comes together:
- Scoring/Calibration compares your eight competency scores to the persona's ground-truth key — your calibration gap per competency (lenient/on/severe) and where notes were inference.
- Feedback Coach gives 1–3 prioritized, concrete improvements.
- You make the call: choose one category and write a one-sentence evidence-based rationale — Ready to Hire Hire with Targeted Ramp Support Hold / Needs More Evidence Do Not Move Forward. The AI checks whether you applied the gates correctly (a professionalism/safety red flag overrides a high average).
Capstone result
The Final Readiness Evaluator returns a capstone readiness band and targeted development guidance:
- Composite (1–5) across the eight competencies + your decision quality.
- Band (Expert 4.5–5.0 · Strong 3.5–4.4 · Interview-Ready 3.0–3.4 · Developing <3.0).
- Per-competency profile, your calibration tendency, top 2 strengths, top 2 focus areas, and a concrete next step.
- Any development flags (prohibited question; unsafe disposition scored as acceptable) as required reinforcement — not a fail.
"Because the capstone exercises all eight competencies plus a live decision, a clean run here is strong evidence for the Strong or Expert band. A rough run just names exactly what to practice next — no penalty for trying."