Senior Data Scientist

What You'll Do

Are you a senior-level data scientist with a passion for building and deploying high-impact fraud or credit risk models? Koalafi is seeking an experienced Data Scientist to lead the development, deployment, and monitoring of machine learning models that sit at the core of our portfolio's profitability. This role requires someone who thrives in an end-to-end environment—designing predictive models, operationalizing them in production, and ensuring they continue to perform in a dynamic market.

You will be a key contributor to Koalafi's decisioning ecosystem, owning models that directly influence credit outcomes, fraud mitigation, and the financial performance of the company. Beyond technical expertise, you will bring strong business intuition, enabling you to translate modeling insights into strategic decisions. This position reports to the Manager of Data Science and regularly partners with senior leaders across Risk, Fraud, Analytics, and Technology.

Responsibilities

• Build, deploy, and maintain production-grade credit and fraud models that are foundational to our real-time decisioning platform and essential to portfolio profitability
• Own the full MLOps lifecycle—from feature engineering, model training, and experiment management to production deployment, performance monitoring, drift detection, and continuous optimization
• Architect and scale end-to-end ML pipelines, ensuring reliability, reproducibility, and seamless integration with core decisioning services
• Design robust model monitoring frameworks that enable tracing, profiling, explainability, and rapid root-cause analysis for production incidents or model degradation
• Partner with data science, risk, and engineering leaders to shape modeling strategy, improve credit policy, and strengthen fraud defenses in response to customer behavior and macroeconomic trends
• Drive continuous improvement of existing models, incorporating new data sources, advanced techniques, and rigorous validation processes
• Communicate complex model logic and insights to non-technical stakeholders, clearly linking modeling decisions to business outcomes and strategic priorities

About You

• 5+ years of hands-on experience building and deploying machine learning models, with a strong grasp of the end-to-end modeling lifecycle from feature engineering to validation and productionization
• 5+ years of professional experience writing performant, maintainable Python code in a collaborative production environment, leveraging core data science libraries like pandas, numpy, xgboost, and scikit-learn
• 2+ Years of experience working on Credit or Fraud risk models
• Proficient in SQL for querying, transforming, and analyzing large datasets, and comfortable working across relational databases and cloud-based data platforms
• Strong understanding of data structures, algorithms, and software engineering principles, and apply them to build robust and scalable data solutions
• Bachelor's degree in a quantitative or STEM field (e.g., Statistics, Mathematics, Computer Science, Engineering) and demonstrate strong analytical and problem-solving skills in your work
• Location Requirement: This position requires regular in-person attendance at one of our two office locations (Richmond, VA or Arlington, VA). Candidates must already be located within a commutable distance to either location, as relocation assistance is not available at this time

Preferred Qualifications

• Advanced technical and analytical background, ideally with a Master's or PhD in a quantitative or STEM field, and a strong understanding of probability, statistics, and predictive modeling algorithms (e.g., Boosting, Random Forests, Decision Trees, Bayesian models)
• Exposure to data and compute platforms such as Snowflake and Databricks
• Background in financial services or experience working in fast-moving, high-growth environments such as startups
• Experience with modern ML infrastructure and tooling, including MLOps frameworks (e.g., MLflow, BentoML), CI/CD automation, and model observability, monitoring, and lifecycle management
• Familiarity with large language models (LLMs) and their deployment in production environments