Senior Data Scientist - Fraud Data Infrastructure & Automation

Socure Socure · Vertical AI · United States · Remote · Data Science & AI

Senior Data Scientist role focused on building agentic AI and LLM-powered systems for fraud detection and identity verification. The role involves designing data pipelines, building and optimizing models with diverse data types, and conducting research to advance fraud detection techniques. It emphasizes end-to-end ownership of projects and collaboration with cross-functional teams.

What you'd actually do

  1. Design, build, and maintain scalable data pipelines and workflows to support analytics, fraud detection, model development, and ongoing data monitoring (e.g., using Spark, Airflow, or similar distributed systems).
  2. Leverage and build agentic AI and LLM-powered systems to automate data exploration, anomaly detection, vendor evaluation, and investigative workflows, increasing the speed and depth of insight generation.
  3. Build and optimize models using a variety of input data types, including tabular data, natural language, point clouds, and images, in support of fraud detection and identity verification use cases.
  4. Own data quality and integrity for critical datasets, implementing monitoring, validation checks, and anomaly detection to ensure reliable input to models and downstream decision systems.
  5. Take ownership of project outcomes from scoping through delivery, managing data quality, technical trade-offs, and timelines; proactively escalate risks and work cross-functionally to resolve challenges.

Skills

Required

  • Python
  • SQL
  • PyTorch
  • TensorFlow
  • scikit-learn
  • Spark
  • Airflow
  • Databricks
  • machine learning algorithms
  • model evaluation techniques
  • data pipeline development
  • fraud prevention
  • risk modeling
  • identity verification

Nice to have

  • LLMs
  • agentic frameworks
  • point clouds
  • images

What the JD emphasized

  • agentic AI
  • LLM-powered systems
  • fraud detection
  • identity verification
  • data pipelines
  • model development
  • novel algorithms

Other signals

  • Leverage and build agentic AI and LLM-powered systems
  • Build and optimize models using a variety of input data types, including tabular data, natural language, point clouds, and images
  • Conduct in-depth research to explore new data sources and develop novel algorithms and features that advance the state of the art in fraud detection, identity resolution, and risk scoring.