Interview scorecard with calibration rubric that produces hire decisions people actually trust

# Interview Scorecard — Senior Backend Engineer **Role:** Senior Backend Engineer **Level:** L5 (5–8 years experience) **Competencies:** Technical depth, system design, collaboration, ownership, code quality --- ## Rating Scale — Behavioral Anchors ### Technical Depth - **1 — Surface only.** Can describe what they have used, not how or why. Struggles with follow-up questions below the tutorial level. - **2 — Working knowledge.** Can explain tradeoffs for tools they have used recently. Gets vague outside their direct experience. - **3 — Solid depth.** Reasons from first principles on topics in their stack. Gives concrete examples of decisions they made and revisited. - **4 — Expert signal.** Teaches the interviewer something. Discusses failure modes and tradeoffs across multiple systems they have shipped. **Sample question:** "Walk me through how you would design idempotency for a payment processor. What breaks first under load?" --- ### System Design - **1** — Produces a block diagram without justification. No discussion of scale, consistency, or failure. - **2** — Covers happy path, misses failure modes. Does not quantify load assumptions. - **3** — States assumptions, sketches trade-offs, identifies 2–3 failure modes with mitigations. - **4** — Proactively explores 3+ alternatives, justifies one, and names what they would monitor in production. **Sample question:** "Design a rate limiter for an API serving 100k req/sec across 4 regions." --- ### Collaboration - **3** — Describes specific conflicts with coworkers and the resolution. Credits teammates concretely. - **4** — Shows evidence of changing their mind based on a colleague's input. Discusses growth from feedback. **Sample question:** "Tell me about a time you disagreed strongly with a peer's technical direction." --- ### Ownership - **3** — Owns outcomes beyond their ticket. Describes following problems across team boundaries. - **4** — Has proactively fixed things outside their mandate and can articulate why it mattered. --- ### Code Quality - **3** — Writes clean code under time pressure. Adds tests without prompting. - **4** — Refactors as they go. Explains naming choices. Catches their own edge cases. --- ## Calibration Rules - A **3** is the bar. Most hires should be mostly 3s with a 4 in at least one competency. - A **4** requires evidence the candidate has operated at the next level already. - Average score alone is never the decision — patterns matter more than means. --- ## Auto-Disqualifying Red Flags - Disparaging former teammates without self-reflection - Inability to describe any past decision they would reverse - Cannot write working code in their preferred language during a live exercise - Evidence of dishonesty on resume claims when probed --- ## Final Recommendation - **Strong Hire** — At least one 4, no score below 3, unique addition to the team - **Hire** — All 3s, no red flags, fills a specific need - **No Hire** — Any 2 in a core competency, or 3s that feel generous - **Strong No Hire** — Any 1, any red flag, or unanimous hesitation from panel