Data Scientist II: Big Data R&D & Identity Graph

Socure; San Francisco (CA); Hybrid; USD 150,000 - 190,000; Full time

Job summary

Socure is building identity trust infrastructure for the digital economy. The role focuses on graph-based ML and large-scale data pipelines to improve deceased monitoring and compliance products, partnering with data scientists and engineers.

You will develop features, evaluate new data sources, and support data processing with Spark, PySpark, and AWS, while communicating insights to cross-functional partners.

Qualifications

• Advanced degree in data science, analytics, or equivalent practical experience.
• Strong Python/Scala programming and SQL proficiency.
• Experience with large-scale ML, graph analytics, and data pipelines.

Responsibilities

• Design and implement machine learning, data mining, and graph-based algorithms for very large datasets.
• Refine entity-resolution and identity-matching algorithms for verification and compliance.
• Build and maintain ETL pipelines using Spark/PySpark and AWS.
• Support feature engineering, data exploration, and A/B testing for models.
• Evaluate new data sources and data quality, summarize impact on model performance.
• Write and optimize SQL/Python/R code for data extraction, transformation, and validation.

Skills

Python

Scala

SQL

ML fundamentals

Problem solving

Communication

Education

Master’s degree in data science or analytics

Ph.D. in data science or analytics (optional)

Tools

Spark/PySpark

scikit-learn

XGBoost

TensorFlow/PyTorch

AWS (EMR, S3)

Databricks

Neo4j / AWS Neptune

GraphFrames

Airflow

Job description

Socure is building identity trust infrastructure for the digital economy. The role focuses on graph-based ML and large-scale data pipelines to improve deceased monitoring and compliance products, partnering with data scientists and engineers.

You will develop features, evaluate new data sources, and support data processing with Spark, PySpark, and AWS, while communicating insights to cross-functional partners.