AI Subrogation Recovery Opportunity Scorer

Quilvex Systems

Remote · $100k – $130k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

Subrogation is the insurance practice of recovering claim payments from liable third parties. A homeowner's insurance company pays a water damage claim, then discovers the damage was caused by a contractor who installed faulty plumbing. The insurer can recover the claim cost from the contractor's liability carrier. This recovery improves combined ratios and directly affects insurer profitability. The problem is identification: out of 10,000 closed claims per month, a subrogation team of 12 investigators can actively pursue maybe 400. Which 400? Currently the decision is made by supervisors scanning claim notes manually. A scoring model that ranks claims by recovery probability and estimated recovery value would triple the team's effective throughput. You'll build the subrogation opportunity scoring system: engineering features from claim notes, liability indicators, third-party identification signals, and historical recovery outcomes, training the scoring model, and integrating the ranked claim queue into the subrogation team's workflow. The ROI is directly measurable: recovery dollars per investigator-hour before and after model deployment.

Fully remote.

Responsibilities

Build subrogation opportunity scoring model ranking closed claims by recovery probability and value
Engineer features from claim notes, liability indicators, and third-party identification signals
Implement NLP extraction identifying third-party fault indicators in unstructured claim notes
Create recovery value estimation model predicting expected recovery amount per claim
Integrate scored claim queue into subrogation team workflow and case management system
Monitor model performance against realized recovery outcomes and retrain on new data

Requirements

3+ years in insurance analytics, claims data science, or financial services ML
Strong Python, SQL, and scikit-learn skills for structured data modeling
Experience with NLP extraction from short-form text like notes or descriptions
Understanding of insurance claim processes and liability concepts preferred
Ability to present model outputs in terms meaningful to insurance operations teams