Lead Decision Scientist

About LemFi: Building the Future of Global Finance for Immigrants

We are LemFi, a dynamic Fintech company committed to transforming how the global diaspora community across North America, Europe, and the United Kingdom manages and moves their money. (YC S21, Series A). Our mission is to build innovative products and services that empower our customers to effortlessly send, receive, manage, and do more with their finances, all within a single, intuitive app. With a rapidly growing community exceeding 1 Million customers, we invite you to join us in shaping the future of inclusive financial services worldwide.

Who Thrives at LemFi?

You are a driven individual who genuinely thrives in the fast-paced, collaborative environment of a scaling Fintech startup. Our culture values humility, upliftment, and a strong work ethic. You are passionate about your area of expertise, yet you readily step up and lend your skills across teams to ensure the success of the company and, most importantly, the satisfaction of our diverse clientele.

The Opportunity: Lead Decision Scientist, Credit & Risk

We are actively seeking a highly analytical and strategically-minded Lead Decision Scientist to champion the development, deployment, and continuous optimization of our credit decisioning and risk modeling frameworks. In this pivotal role, you will directly influence our lending strategy, build essential data products, and significantly impact portfolio performance through insightful, data-led analysis. This position is perfect for someone with deep technical and analytical capabilities who is passionate about data exploration, building robust models, and driving improvements through experimentation.

Your Impact: What You’ll Lead

As our Lead Decision Scientist, you will be instrumental in building and scaling our lending capabilities. Your key areas of contribution and ownership will include:

Developing and Enhancing Models

Leading the end-to-end lifecycle of credit risk and affordability models, from strategic conception and data sourcing (including bureau, open banking, and alternative behavioral data) through feature engineering, development, rigorous validation, and ongoing performance monitoring.

Driving Optimization Through Experimentation

Designing and executing champion/challenger tests and A/B experiments to iteratively enhance key metrics such as approval rates, minimize loss rates, and improve the overall customer lending experience.

Generating Actionable Insights

Deeply analyzing credit portfolio performance data to uncover critical insights that directly inform strategic business decisions.

Building and Mentoring a Team

Guiding and developing a growing team of analysts and data scientists as the department scales, fostering a culture of technical excellence and collaboration.

Collaborating Across Functions

Partnering closely with the Data Engineering team to seamlessly deploy validated models into production pipelines and working hand-in-hand with various business stakeholders to define clear modeling objectives and effectively communicate complex model outcomes in a business context.

What You Bring: Skills & Experience

To succeed in this role, you should possess a strong foundation and proven track record:

Core Requirements

Demonstrated experience of 5-7 years in the consumer credit domain, with significant focus on data science or decision science methodologies.

Extensive hands-on experience building and deploying machine learning models using Python, leveraging key libraries such as scikit-learn, XGBoost, or LightGBM.

Solid expertise working with complex transactional datasets (like Open Banking and Categorisation data) and traditional bureau data (e.g., Experian, Equifax).

Deep understanding of feature engineering techniques, data preprocessing, and effective strategies for handling class imbalance in modeling.

Proficiency in evaluating model performance using appropriate metrics (e.g., AUC, KS, precision/recall) and performing thorough validation across diverse customer segments.

Familiarity with standard best practices for model monitoring, performance tracking, and detecting data drift in production environments.

Strong command of SQL for efficient data extraction, joining, and transformation from various sources.

Extra Edge: Highly Valued Skills

Experience in these areas would be a significant asset:

Familiarity with unsupervised learning techniques (e.g., K-means, DBSCAN, PCA, autoencoders) and their practical applications in credit use cases like behavioral segmentation or exploratory analysis.

Experience working within a fast-paced startup or scale-up environment, accustomed to rapid decision-making cycles.

Exposure to leveraging alternative data sources (such as device data or psychometric scoring) for building credit scoring models.

“