Manu's Vault

Humane-Interview-Process

4 min readStatus:Seedling

Most engineering interview processes are simultaneously burdensome for candidates and inaccurate as selection mechanisms. Larson’s humane interview process framework addresses both problems at once — the practices that reduce candidate burden typically also improve predictive validity.

The Seven Principles

1. Interview for the actual job

  • Evaluate candidates on work resembling the real role, not abstract puzzles
  • A staff engineer doing 90% system design should be assessed on system design, not linked-list reversal
  • Failure mode: whiteboard algorithmic challenges for roles that never require them

2. Avoid gotcha questions

  • Questions designed to catch candidates out test memorised trivia, not job-relevant judgment
  • Replace gotchas with open-ended questions that reveal how candidates reason
  • Failure mode: “What’s the difference between a mutex and a semaphore?” when the team uses higher-level concurrency abstractions

3. Use a consistent scoring rubric

  • Pre-define evaluation criteria and a scoring framework before interviews begin
  • Without rubrics, interviewers default to “would I enjoy working with this person?” — a bias-prone proxy
  • Failure mode: free-form written feedback where every interviewer assesses different dimensions

4. Calibrate interviewers

  • New interviewers shadow experienced ones before solo interviews
  • Regular calibration sessions compare scores on borderline candidates to surface drift
  • Failure mode: interviewers who have never compared their scoring to others’ on the same candidate

5. Keep it short

  • Four to five interviews captures most available signal; six or more returns diminishing information
  • Multi-day on-sites filter for endurance, not job fit; candidates are often using PTO
  • Failure mode: seven-round loops designed to achieve consensus rather than accuracy

6. Reduce performance anxiety

  • Whiteboard coding under observation introduces cognitive load unrelated to job performance
  • Take-home exercises, pair programming sessions, or design discussions reduce the anxiety component
  • Anxiety in interviews ≠ anxiety on the job
  • Failure mode: standardising on the format that is easiest for interviewers to run, not easiest for candidates to perform in

7. Collect evidence, not opinions

  • Feedback should read: “candidate explained the trade-off between X and Y, demonstrating understanding of Z”
  • Not: “candidate seems smart” or “I liked their energy”
  • Failure mode: adjective-heavy feedback that cannot be audited for bias or inconsistency

The Implicit Principle

Fairness and accuracy align. A process that respects candidate time and reduces anxiety tends to elicit more authentic performance, which improves predictive validity. Optimising for candidate experience is not a trade-off with hiring quality — it reinforces it.

Connection to the Hiring Funnel

The humane interview process governs the Evaluate stage of the Hiring-Funnel. Poor evaluation design degrades the entire funnel: sourcing effort from Three-Candidate-Sources and cold outreach is wasted if the interview process filters for the wrong signals or drives candidates to withdraw.

Sources

  • Schmidt, Frank L. and John E. Hunter (1998). “The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings.” Psychological Bulletin, Vol. 124, No. 2, pp. 262–274. DOI: 10.1037/0033-2909.124.2.262.

    • Definitive meta-analysis showing structured interviews have substantially higher predictive validity (~0.51) than unstructured interviews (~0.38) for job performance

  • Larson, Will (2019). An Elegant Puzzle: Systems of Engineering Management. Stripe Press. ISBN: 978-1-7322651-8-9.

    • Chapter 6.2: Interview process design principles

  • Campion, Michael A., David K. Palmer, and James E. Campion (1997). “A Review of Structure in the Selection Interview.” Personnel Psychology, Vol. 50, No. 3, pp. 655–702. DOI: 10.1111/j.1744-6570.1997.tb00709.x.

    • Identifies fifteen dimensions of interview structure and their impact on reliability and validity; foundational framework for rubric design

  • Bohnet, Iris (2016). What Works: Gender Equality by Design. Harvard University Press. ISBN: 978-0-674-08903-3.

    • Chapter 6 addresses how structured evaluation criteria reduce affinity bias in hiring; empirical evidence from field experiments

  • Macan, Therese Hoff (1994). “Time Management: Test of a Process Model.” Journal of Applied Psychology, Vol. 79, No. 3, pp. 381–391.

    • Cognitive load research supporting the case for reducing extraneous stressors in evaluation contexts; basis for anxiety-reduction principle

Note

This content was drafted with assistance from AI tools for research, organization, and initial content generation. All final content has been reviewed, fact-checked, and edited by the author to ensure accuracy and alignment with the author’s intentions and perspective.

Graph View

Backlinks