Experience: 8+ years of experience as a Data Scientist, with a proven track record in solving complex problems, particularly in fraud detection or financial crime.
Machine Learning Expertise: Extensive experience in designing, developing, and deploying machine learning models to detect and mitigate fraud. You should be comfortable translating business challenges into data-driven solutions.
Working with Large-Scale Data: Proficiency in handling large, tabular datasets, and applying robust techniques for data analysis and model training.
Advanced Tools and Platforms: Experience with tools such as PySpark, Databricks, AWS, or GCP for processing large datasets, training models, and deploying them at scale.
Production-Ready Models: Proven ability to deploy models into production environments, optimizing them for performance and scalability, while ensuring they remain effective over time.
Data & Model Observability: Expertise in monitoring and maintaining the health and performance of models post-deployment to ensure continuous improvement and fraud detection accuracy.
Fintech & Fraud Detection: Background in the Fintech industry, with specific experience in financial crime and fraud detection, applying data science to solve real-world business problems.
Collaboration & Communication: Strong interpersonal skills to collaborate effectively with data engineers, machine learning engineers, and product managers in an agile, iterative environment. Ability to communicate complex insights clearly to both technical and non-technical stakeholders.
Responsibilities:
Collaborating with the FinCrime team: Work closely to identify and solve fraud detection problems, using data science to drive business decisions and significantly reduce fraudulent activities.
Translating business requirements: Understand business needs and translate them into data products and models that address specific fraud detection challenges.
Building and optimizing models: Train machine learning models, optimize hyperparameters, design KPIs, and implement experiments to improve fraud detection accuracy and business outcomes.
Productionizing models: Work with machine learning engineers and data engineers to deploy models into production, ensuring they are scalable and optimized for real-time fraud detection.
Adopting new methodologies: Lead the adoption of innovative methods and technologies, continuously improving fraud detection models and data science practices.
Coaching junior data scientists: Mentor and guide junior team members, setting best practices for model development, optimization, and deployment.
Being a technical subject matter expert: Serve as a subject matter expert, providing guidance on complex technical concepts related to fraud detection, machine learning, and data science.