What we’re looking for

M1’s Risk Analytics team is looking for a Data Scientist to assist in analysis and model development as well as deploying and maintaining code in our loan decisioning system. This position will primarily focus on our lending s products as well as analysis and modeling related to general fraud risk.

The candidate will have the following or will have demonstrated the ability and willingness to learn the following:

• A familiarity with financial mathematics. For example, you are familiar with present value calculations. You can understand and create a simple loan amortization table.
• A strong familiarity with statistics and statistical analysis. For example, you are comfortable creating and interpreting linear regression models, or you are familiar with A/B testing and basic experiment design.
• A strong familiarity with Machine Learning and predictive model development.
• Production-level python coding and development skills. You can write clean, efficient code, unit test your code, and work with engineers to deploy the code in production environments.
• An ability to grapple with complex business problems, interpret them into analytical problems, design build a solution, and work with cross-functional teams to put that solution to work. For instance, you can analyze fraud data and understand when a machine learning solution is warranted or when an analysis-informed rule set would suffice based on your measure of potential impact.
• Writing skills sufficient to document models and processes for both internal stakeholders and external regulators and auditors.
• Experience working with direct mailing campaigns, preferably in a financial services industry

The Data Scientist is someone who has a deep understanding of Advanced Analytics and Machine Learning. The role will include work on descriptive analytics as well as building predictive/prescriptive models and deploying those models into production. A successful Data Scientist has demonstrated experience understanding business problems, identifying where models can help solve those problems, building models, and working with teams to make sure those models have a high impact.

What You'll Do

This position is within M1’s Risk Analytics team and focuses on credit risk and loan pricing as well as fraud identification, prevention, and mitigation. This role will also assist with analyzing and informing direct mail campaigns for lending products.

In this role you will:

• Help determine loan pricing and underwriting strategies to balance growth and profitability:
• Analyze the performance of past loan cohorts to estimate current and projected portfolio profitability. Analyze pricing relative to benchmarks and competitors. Help communicate results to stakeholders.
• Improve M1’s loan pricing model using a combination of analysis-informed rule changes and predictive modeling.
• Write python code to implement rule changes and models into production.
• Write and maintain model documentation for approval by internal and external stakeholders.
• Help create and maintain rules and models to improve fraud screening efficiency or reduce fraud losses.
• Work with data analysts and fraud teams to understand types of fraud, how they are measured, and what interventions could improve efficiency.
• Create and implement ML models to improve fraud detection and mitigation interventions.
• Help improve direct mail campaign performance for our lending products:
• Review analysis from data analysts and vendors on campaign performance.
• Advise on A/B testing and experiment design strategies to improve audience selection and creative design.
• Work closely with software engineers and data engineers on production deployment and data pipelines.
• Work with compliance and accounting teams to ensure adherence to reporting and documentation requirements.
• Have the opportunity to present analysis, findings and proposals to leadership.

Qualifications

• Master’s degree in a quantitative field (like Economics, Mathematics, Statistics or Data Science) and at least two years of experience in a technical role. Alternately, a Master’s degree in a technical field (like Computer Science, Data Science) and at least three years of experience in a heavily analytical role requiring advanced analytics and strong grasp of statistics.
• In lieu of a Master’s degree, a Bachelor’s degree in a quantitative/technical field and at least five years of experience in a relevant role demonstrating both quantitative and technical abilities.
• Three years of experience using advanced analytics or predictive modeling to solve real-world business problems.
• Demonstrated experience writing high-quality python code.
• Demonstrated experience with a ML coding framework (e.g. scikit-learn) or auto-ML tool (e.g. Sage Maker, Dataiku).
• Deep, demonstrated experience applying any of the following methods in real-world applications: causal inference, Bayesian data analysis, statistical experiment design, complex regression modelling and analysis, survival analysis.
• Ability to explain complicated technical concepts to non-technical stakeholders.
• Eagerness to learn new skills, tools, and subjects.