The latest articles on DEV Community by Devesh Pareek (@decipheringlogic). https://dev.to/decipheringlogic https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4108446%2Fd12d9b2a-13d0-40fe-b260-dac5470e0d29.png https://dev.to/decipheringlogic en Devesh Pareek Wed, 16 Sep 2026 11:42:07 +0000 https://dev.to/decipheringlogic/building-a-candidate-readiness-score-for-a-healthcare-staffing-marketplace-5ac2 https://dev.to/decipheringlogic/building-a-candidate-readiness-score-for-a-healthcare-staffing-marketplace-5ac2 <p>Wanderly is a travel healthcare staffing marketplace: 663,000+ registered candidates, 500+ agencies. Recruiters there had one recurring question: of the thousands of candidates in my list, who do I call first?</p> <p>We built a Readiness Score to answer it. This is the version-one model, the reasoning behind its shape, and what we deliberately left out.</p> <p>## Three signals, not eight</p> <p>The full design had eight signals: recency, job views, clicks, applications, email and SMS engagement, an opt-out penalty, campaign participation, profile completeness. It was a good model. We shipped three:</p> <ul> <li>Campaign engagement, 50%</li> <li>Application activity, 30%</li> <li>Profile completeness, 20%</li> </ul> <p>Fewer signals meant the score was explainable to a recruiter in one sentence, and it meant we could validate the model against sixty days of real pilot data before adding anything.</p> <p>## Recency decay</p> <p>Campaign engagement is decayed by recency: activity in the last 30 days counts at 1.0x, 31 to 90 days at 0.7x, 91 to 180 days at 0.4x, and nothing beyond the configured window. A reply from last week and a reply from five months ago are not the same signal.</p> <p>## Hard overrides</p> <p>Some things are switches, not inputs.</p> <ul> <li>Opt-out (a STOP reply): score forced to 0 and the candidate is removed from every campaign audience, regardless of any other signal.</li> <li>Twelve months of total inactivity: score capped at 25.</li> <li>New candidate with no history: scored on profile completeness only, shown with a New badge.</li> </ul> <p>Compliance sits outside the model. A weighted penalty can be outvoted by other signals; a hard override cannot.</p> <p>## Tiers</p> <p>Hot 70 to 100. Warm 40 to 69. Cold 20 to 39. Inactive 0 to 19. Minimum campaign-eligible score: 20. Weights are locked for v1; admins configure the time windows and the tier thresholds.</p> <p>## What the recruiter sees</p> <p>The number, a colour, and a breakdown: Campaign engagement 42/50, Application activity 24/30, Profile 18/20, based on the last 6 months. Opted-out candidates are flagged, not hidden.</p> <p>## What we left for v2</p> <p>Behavioural events (job views, time on job detail, pay comparisons, saves), agency-specific weights, and a second scoring mode for agencies that use the marketplace without the campaign product. All of it waits on real data from v1.</p> <p>Originally published at decipheringlogic.com</p> ai rag vectordatabase Devesh Pareek Wed, 16 Sep 2026 10:18:02 +0000 https://dev.to/decipheringlogic/wanderly-case-study-3b74 https://dev.to/decipheringlogic/wanderly-case-study-3b74 <h1> Building a candidate Readiness Score for a healthcare staffing marketplace </h1> <p>Wanderly is a travel healthcare staffing marketplace: 663,000+ registered candidates, 500+ agencies. Recruiters there had one recurring question: of the thousands of candidates in my list, who do I call first?</p> <p>We built a Readiness Score to answer it. This is the version-one model, the reasoning behind its shape, and what we deliberately left out.</p> <p>## Three signals, not eight</p> <p>The full design had eight signals: recency, job views, clicks, applications, email and SMS engagement, an opt-out penalty, campaign participation, profile completeness. It was a good model. We shipped three:</p> <ul> <li>Campaign engagement, 50%</li> <li>Application activity, 30%</li> <li>Profile completeness, 20%</li> </ul> <p>Fewer signals meant the score was explainable to a recruiter in one sentence, and it meant we could validate the model against sixty days of real pilot data before adding anything.</p> <p>## Recency decay</p> <p>Campaign engagement is decayed by recency: activity in the last 30 days counts at 1.0x, 31 to 90 days at 0.7x, 91 to 180 days at 0.4x, and nothing beyond the configured window. A reply from last week and a reply from five months ago are not the same signal.</p> <p>## Hard overrides</p> <p>Some things are switches, not inputs.</p> <ul> <li>Opt-out (a STOP reply): score forced to 0 and the candidate is removed from every campaign audience, regardless of any other signal.</li> <li>Twelve months of total inactivity: score capped at 25.</li> <li>New candidate with no history: scored on profile completeness only, shown with a New badge.</li> </ul> <p>Compliance sits outside the model. A weighted penalty can be outvoted by other signals; a hard override cannot.</p> <p>## Tiers</p> <p>Hot 70 to 100. Warm 40 to 69. Cold 20 to 39. Inactive 0 to 19. Minimum campaign-eligible score: 20. Weights are locked for v1; admins configure the time windows and the tier thresholds.</p> <p>## What the recruiter sees</p> <p>The number, a colour, and a breakdown: Campaign engagement 42/50, Application activity 24/30, Profile 18/20, based on the last 6 months. Opted-out candidates are flagged, not hidden.</p> <p>## What we left for v2</p> <p>Behavioural events (job views, time on job detail, pay comparisons, saves), agency-specific weights, and a second scoring mode for agencies that use the marketplace without the campaign product. All of it waits on real data from v1.</p> <p>Originally published at decipheringlogic.com</p> startup productdevelopment mvp