Introducing bidirectional scoring: how Chesky awards and penalizes
Most evaluation tools only add points. Chesky's scoring engine also deducts — because honest assessment means acknowledging gaps, not just celebrating strengths.
Every evaluation tool we looked at before building Chesky worked the same way: start from zero and add points for good signals. A strong LinkedIn? +10. Relevant GitHub contributions? +8. Years of experience? +5. Add it all up, and you get a score.
This approach has a fundamental flaw: it ignores the negative. A candidate with a strong LinkedIn and a pattern of 6-month job tenures over 8 years isn't the same as a candidate with a strong LinkedIn and a stable 4-year track record. A score that only adds will give them the same number.
How bidirectional scoring works
Chesky's scoring engine evaluates each candidate against the specific job description on six dimensions. Within each dimension, the engine identifies both positive evidence (signals that increase the score) and negative evidence (signals that decrease it).
Positive evidence: demonstrated competency, relevant experience, verifiable achievements, consistent trajectory, cultural alignment signals.
Negative evidence: unexplained gaps, pattern of short tenures without context, inconsistencies between claims and verifiable data, AI-generated case study content, absence of expected competency signals for the seniority level.
The result is a 0–100 score that reflects the actual balance of evidence — not just an accumulation of positives. A candidate who looks good on paper but raises red flags gets a score that reflects both dimensions.
Why this matters for defensibility
When a hiring committee reviews two candidates with scores of 87 and 61, they can ask: where did the points go? Chesky shows you exactly — which dimension scored high, which scored low, and what specific evidence drove each. The decision becomes auditable by design.