Interview Types

Current: Behavioral Only

All interviews currently use behavioral questioning — "Tell me about a time when..." This is the baseline interview type.

Planned: Situational (Phase 10)

"What would you do if..." hypothetical questions. Requires different scoring approach:

• Cannot use STAR/behavioral anchors on hypotheticals
• Evaluates reasoning quality, not past experience
• Different rubric types: situational_interviewer, situational_scorer

Your input needed:

• What makes a good vs. poor hypothetical answer?
• Define scoring anchors for hypothetical responses
• Specify prompt instructions enforcing valid situational technique
• Estimated 1-2 weeks of design

Planned: Technical (Phase 11)

Three tiers:

1. Conversational technical — verbal problem-solving ("Walk me through how you'd design..."). Same Retell infrastructure.
2. Live coding — Monaco editor embedded in text interview. AI sees code as text in transcript.
3. External platforms — LeetCode/HackerRank/CodeSignal integration via webhooks.

Key decisions already made (by Dr. Rigos):

• Rubrics stay system-wide (not per-discipline)
• Practical scoring uses regular competency-based scoring
• Disciplines are organizational only (workspace type, difficulty level), not scoring constructs

Your input needed:

• Define "technical competence" psychometrically (content validity evidence)
• Design scoring rubrics distinguishing knowledge from communication skill
• Establish "interview" vs "test" legal boundary
• Review for construct contamination
• Estimated 2-4 weeks per domain

Planned: Post-Hire (Phase 12)

Stay interviews, exit interviews, onboarding check-ins, engagement pulses. Different personas, thematic analysis instead of 1-5 scoring.

Your input needed:

• Design question spines per context
• Define insight extraction rubric (themes, sentiment, risk flags)
• Specify claim boundaries (qualitative insights, not scores)