Data Scientist (Fintech)

TL;DR

Data Scientist (Fraud Detection): Designing and improving fraud detection systems for motor insurance across UK, Spain, and Italy markets with an accent on prioritizing high-risk claims and operational efficiency. Focus on identifying emerging fraud patterns, analyzing suspicious networks, and delivering measurable business impact through applied machine learning.

Location: Hybrid in Milan, Italy

Company

A leading online motor insurance provider operating in Italy, UK, and Spain, utilizing data and tech to rethink the insurance experience.

What you will do

• Design and optimize fraud detection systems to balance detection accuracy with operational efficiency.
• Collaborate with fraud analysts, engineers, and product managers to improve detection capabilities.
• Identify emerging fraud patterns and anomalous behavioral patterns across the insurance lifecycle.
• Analyze suspicious networks of related claims and synthetic entities to prevent fraud.
• Translate data insights into effective detection and prevention strategies that impact the company's bottom line.

Requirements

• 3+ years of experience in data science or applied machine learning, leading end-to-end initiatives.
• Strong Python programming skills and fluency with Git.
• Proficiency in SQL, Spark, and standard data libraries (Pandas, Sklearn, Keras, XGBoost, LightGBM).
• Strong analytical and quantitative mindset to guide decision-making with evidence.
• Must be based in or able to work hybridly in Milan, Italy.

Nice to have

• Experience in analytical fraud detection within Insurance, Finance, or Law Enforcement domains.
• In-depth knowledge of graph databases (Neo4j, AWS Neptune) and analytics (NetworkX, GNN).
• Experience with BI tools such as Tableau.

Culture & Benefits

• Hybrid working model combining home and office days.
• Flexibility to work from anywhere for up to 30 days per year.
• Private healthcare, gym discounts, and mental health support programs.
• Access to mentorship, learning resources, and personalized growth plans.