Senior Data Scientist (International eKYC, Identity Graph)

Role Overview

Senior Data Scientist (International eKYC, Identity Graph) at Socure. You will develop global identity verification solutions with a focus on international eKYC, designing and deploying machine learning and graph-based systems for diverse markets and regulations.

Responsibilities

• Conception et déploiement of machine learning and graph-based systems adapted to international markets.
• Own complex, cross-functional initiatives such as:
  • International identity graph evolution
  • Probabilistic matching (probabilistic reconciliation)
• Partner closely with Product, Engineering, Compliance, and GTM teams to launch and scale eKYC solutions across multiple countries and regions.
• Build and productionize models and pipelines in a high-stakes domain.

Requirements

• 6+ years hands-on applied ML/Data Science experience (4+ with a Ph.D.), including ownership of production models and pipelines in domains such as fraud, risk, identity, payments, or credit.
• Strong preference for experience on international/multi-region products (e.g., cross-country KYC, credit risk, payments, compliance systems).
• Master’s or Ph.D. in a relevant field (Computer Science, Data Science, Machine Learning, Statistics, Mathematics) or equivalent practical experience.
• Feature engineering for noisy and heterogeneous identity data.
• Proven expertise in graph technologies and graph algorithms, including:
  • Graph/identity concepts such as entity resolution and link prediction
  • Community detection and label propagation
  • Ability to work with graphs at scale
• Expert proficiency in Python and SQL.
• Experience with distributed data processing on very large datasets (e.g., Spark/PySpark, Databricks or similar).
• Robust evaluation under label sparsity and feedback delays.
• Deep experience designing/training/deploying models for:
  • classification, ranking, anomaly detection, and/or graph learning
• Calibration and thresholding tailored to regional risk and regulatory constraints.

Nice-to-Haves

• Experience with specific graph tooling such as Neo4j, AWS Neptune, GraphFrames, DGL, PyTorch Geometric.