Principal Data Scientist

Capital One Capital One · Banking · Bangalore, IN

Capital One's DataLabs AML team is modernizing money laundering detection using advanced analytics, statistics, and machine learning models. This Principal Data Scientist role involves end-to-end development, deployment, and monitoring of risk management models, partnering cross-functionally, and defending these systems to regulators. The role requires strong ML/statistical skills, experience with Python/R, and SQL, with a focus on building and validating predictive models in the financial services domain.

What you'd actually do

  1. Partner cross-functionally with teams throughout the enterprise including Product and Tech to enable capturing risky behavior.
  2. Collaborate with Quants, Data Scientists, Data analysts, Business analysts and risk managers to drive innovation, shape strategic initiatives and propel AML Transaction monitoring to new heights in the data management landscape.
  3. Assess, challenge, and sometimes defend state-of-the-art decision-making systems to internal and regulatory partners.
  4. Build upon your existing machine learning and statistical toolset - both by learning new technologies and by building custom software tools for data exploration, model performance evaluation, and more.
  5. Build data products, data solutions, tools, and capabilities to enable self-service frameworks for data consumers.

Skills

Required

  • Python or R
  • model development or validation
  • relational databases and SQL
  • analytical skills
  • communication skills

Nice to have

  • econometric and statistical techniques (such as predictive modeling, logistic regression, panel data models, decision trees, machine learning methods)
  • financial services industry
  • Fraud / Anti-money Laundering domain

What the JD emphasized

  • defend state-of-the-art decision-making systems to internal and regulatory partners
  • defend the model with validators, auditors and regulators

Other signals

  • develops data sourcing, predictive models, monitoring, and reporting
  • end-to-end development, deployment, and monitoring of these critical risk management models
  • assess, challenge, and sometimes defend state-of-the-art decision-making systems to internal and regulatory partners
  • build custom software tools for data exploration, model performance evaluation, and more
  • build data products, data solutions, tools, and capabilities to enable self-service frameworks for data consumers
  • defend the model with validators, auditors and regulators