Sr Data Scientist

Zepz•,
•Remote

About The Position

Zepz is hiring a Senior Data Scientist to champion data science within its FinCrime-Fraud-AI-ML team. The role focuses initially on fraud detection, with opportunities to expand into risk and credit rating. The successful candidate will design, build, deploy, and own machine learning models that prevent fraud while preserving a positive experience for legitimate customers. The Senior Data Scientist will work in a collaborative FinCrime squad alongside Data Scientists, Machine Learning Engineers, and Backend Engineers, and partner with fraud analysts, operations, product, and data and analytical engineering teams. The position combines hands-on model development and lifecycle ownership with experimentation, commercial decision-making, stakeholder enablement, and technical leadership. Zepz is a remote-first company, with work-from-anywhere flexibility within the employee’s employing country.

Requirements

  • At least 6 years of experience in data science.
  • Hands-on experience with regression analysis, clustering methods, boosting algorithms, feature and model selection, and training and deploying models that deliver measurable value.
  • Experience in fraud, financial crime, risk, or a similar adversarial domain; payments, fintech, or banking experience is ideal.
  • Experience with imbalanced classification and delayed or noisy labels.
  • Strong Python coding skills, including using Python for automation and application deployment.
  • Expert SQL skills, including CTEs, joins, and subqueries for manipulating complex modeling datasets.
  • Ability to deploy models and set up the monitoring they need.
  • Familiarity with modern data science tools and platforms such as scikit-learn, BigQuery, Redshift, or Databricks, and visualization tools such as Looker, Tableau, or Streamlit.
  • Ability to drive technical results within a business-oriented framework and balance technical excellence with commercial impact.
  • Strong problem-solving skills, comfort with over-communication, and adaptability to changing priorities.
  • Curiosity about AI and its applications in the role, plus practical professional experience using AI tools to improve productivity.
  • Strong AI literacy, including effective prompting, critical evaluation of AI-generated output, and understanding of AI assistants' capabilities and limitations.
  • Awareness of ethical considerations and responsible AI use, particularly with sensitive data, customer experience, and compliance.

Nice To Haves

  • Experience with AWS services such as SageMaker, ECS, and EMR.
  • Advanced knowledge of experimentation design and evaluation.
  • Graph analytics or graph machine learning for detecting fraud rings and mule networks.
  • Experience with real-time or streaming machine learning and low-latency scoring.
  • Knowledge of model explainability techniques such as SHAP and experience with model risk governance.
  • Knowledge of cross-border payments or remittances and related fraud and AML regulations.
  • A tenacious approach, including willingness to work manually when needed while seeking opportunities to automate.
  • Prior professional experience using Claude (Anthropic).

Responsibilities

  • Own fraud detection models end to end, from problem framing and feature engineering through training, evaluation, and deployment.
  • Automate model training and deployment, ensuring models are scalable, documented, tested, and monitored for drift.
  • Build models that score transactions, accounts, and customer behaviour in real time to detect threats such as account takeover, stolen payment credentials, first-party fraud, scams, and mule activity.
  • Work with Data and Analytical Engineering to incorporate transactional, device, behavioural, and identity data into models using dbt best practices.
  • Design and run experiments to measure the impact of models, third-party data sources, and emerging tools on fraud losses, approval rates, and customer experience.
  • Translate business needs into data science projects and set thresholds that balance fraud losses against friction for legitimate customers.
  • Continuously improve model accuracy and data quality, adapting quickly as fraud patterns change.
  • Build tools and strategies that help non-technical stakeholders understand and act on model results.
  • Mentor other data scientists, review their work, and help establish technical standards for the team.

Benefits

  • Unlimited annual leave.
  • Healthcare.
  • Employee discounts.
  • Remote-first flexibility to work within the employing country.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service