shiphrahx/ai-for-engineering-leaders/interview-debrief-facilitation

Interview Debrief Facilitation

Produce a debrief plan that forces an evidence-based, anchoring-resistant decision — each interviewer commits independently before the group discusses, and the meeting ends with a documented hire/no-hire.

Inputs to gather

Gather these before writing. If any are missing, ask for them in a single batched question — never invent interviewers, stages, or ratings. Mark anything genuinely unavailable as Unknown in the output.

• Candidate — name
• Role — title and team
• Interviewers and stages — who interviewed and which stage each ran
• Decision rule — the team's existing hire threshold, if one is defined

Steps

1. Define pre-debrief prep: each interviewer submits a written assessment before the meeting (rating on the team's scale, top 2 strengths, top 2 concerns, one memorable signal), and no one reads others' assessments until the meeting — this prevents pre-anchoring.
2. Write the opening (2 min): state the candidate and role, the no-interruption rule, the "don't change your rating based on what you hear yet" rule, and that the hiring manager shares last to avoid anchoring.
3. Structure the round-robin (~3 min each): an order that puts the hiring manager last; each interviewer shares rating, top strength, top concern, one signal — no interruptions.
4. Structure the discussion: ask where the panel agrees, where it disagrees (and force specific evidence on disagreements), what was NOT assessed that matters for this role, and the onboarding risk if hired.
5. Define the decision: a clear rule (e.g. ≥3 of 4 at Hire/Strong Hire with no Strong No Hire), a tiebreak (hiring manager decides on a genuine split), and the requirement to document the decision and rationale in the ATS.
6. Define post-debrief next steps for each outcome (hire → draft offer justification within a set window; no-hire → personalized rejection within a set window).
7. Adapt to context: on strong disagreement (Strong Hire vs Strong No Hire), don't average — dig into what each interviewer actually observed; reinforce anchoring prevention (manager last); for remote debriefs, have everyone paste assessments into a shared doc simultaneously, then discuss verbally; keep it under 30 minutes — an hour of debate is itself a mixed-signal signal.
8. Assemble the output in the format below.

Output format

**Debrief Facilitation Plan**
 
**Pre-Debrief (before the meeting)**
[What each interviewer submits; no one reads others' until the meeting.]
 
**Opening ([N] min)**
"[Candidate, role, rules: no interruptions, hold your rating, manager goes last.]"
 
**Round-robin ([N] min, ~[N] min each)**
Order: [interviewer] → ... → [hiring manager last]
Each shares: rating, top strength, top concern, one signal.
 
**Discussion ([N] min)**
- "Where do we agree?"
- "Where do we disagree? Specific evidence."
- "What didn't we assess that matters for this role?"
- "If hired, what's the onboarding risk?"
 
**Decision ([N] min)**
- [Threshold rule]
- If split: [tiebreak]
- Document decision and rationale in the ATS.
 
**Post-Debrief**
- Hire → [next step + timeframe]
- No hire → [next step + timeframe]

Boundaries

• Never invent interviewer names, stages, or ratings — mark unknowns as Unknown and ask.
• Never average conflicting ratings; surface and discuss the underlying evidence instead.
• Never let the hiring manager share first, and never let one loud voice override documented independent assessments.
• Never base the decision on anything other than role-relevant evidence; do not introduce protected-characteristic considerations.

Chaining

• After a hire decision, offer candidate-evaluation-summary to write up the synthesized evaluation, then offer-justification.
• If panel ratings are inconsistent across candidates, offer interviewer-calibration-guide to align standards before the next loop.