AI Fraud Claims Detection Specialist

Quilvex Systems

Remote · $130k – $160k / 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

Insurance fraud costs the industry $80 billion per year in the US alone. For our clients — regional P&C insurers processing 15,000 claims per month — fraud detection is the difference between a sustainable loss ratio and a company that loses money on every policy. The current detection system flags 3.2% of claims for investigation. Industry data suggests 8-10% of submitted claims involve some form of fraud. We're catching less than half of it. You'll lead the fraud detection model redesign. The problem is multi-layered: first-party fraud, staged accidents, provider billing fraud, and organized ring activity all look different and require different signals. You'll build a multi-model architecture that runs specialized detectors in parallel and aggregates scores through a meta-learner, using features from claims documents, claimant history, network relationships, and external data signals. This is a senior IC role with significant autonomy. You'll define the detection architecture, own the training pipeline, and work directly with the Special Investigations Unit to understand what human investigators actually need from your model outputs.

Fully remote.

Responsibilities

Design multi-model fraud detection architecture for P&C insurance claims
Build specialized detectors for different fraud typologies: first-party, staged accidents, provider fraud
Develop network analysis features capturing relationships between claimants, providers, and attorneys
Create explainability outputs that guide Special Investigations Unit prioritization
Build active learning loops incorporating investigator feedback into training data
Monitor recall against confirmed fraud rates and retrain on emerging patterns

Requirements

6+ years in ML engineering with fraud detection or financial crime domain experience
Experience with graph features and network analysis for fraud ring detection
Strong Python, SQL, and statistical modeling skills
Deep understanding of precision-recall trade-offs in high-stakes classification
Ability to collaborate with investigative domain experts