Lead Machine Learning Engineer – Credit Risk & MLOps
Role Overview
This role is responsible for leading the productionization and ongoing operation of credit risk models within a financial services platform. You will take ownership of transitioning validated models into a secure, production-grade environment, while ensuring continuous improvement through robust MLOps practices.
The position also plays a critical role in implementing responsible AI principles, including model transparency, interpretability, and fairness in lending decisions.
- Key Objectives
- Operationalize machine learning models within a regulated financial system
- Establish reliable CI/CD and MLOps frameworks for model lifecycle management
- Enhance explainability and fairness of credit decision systems
- Scale model training and inference pipelines for high-volume financial data
Core Responsibilities
- ML Pipeline Architecture
- Design and implement end-to-end machine learning workflows
- Integrate data processing and model deployment across cloud-based environments
- Build automated pipelines for training, validation, deployment, and monitoring
- Model Development & Optimization
- Develop and deploy credit scoring models using large-scale datasets
- Optimize performance for both batch and real-time inference scenarios
- Improve model robustness and scalability within production systems
- Explainability & Fairness (Responsible AI)
- Implement interpretability techniques (e.g., SHAP values, counterfactual analysis)
- Evaluate and mitigate bias using quantitative fairness metrics
- Provide clear, auditable explanations for model-driven decisions
- Model Monitoring & Lifecycle Management
- Track model performance and detect degradation (concept drift, data drift)
- Build automated retraining and validation workflows
- Maintain compliance with internal risk and governance standards
- Data Processing & Scalability
- Design efficient pipelines for large transactional datasets
- Utilize distributed processing frameworks for performance optimization
- Improve data handling strategies for structured financial data
Required Qualifications
- Proven experience deploying and maintaining machine learning models in production
- Hands-on expertise across the full ML lifecycle (development → deployment → monitoring)
- Strong programming skills in Python with common ML libraries (e.g., pandas, scikit-learn)
- Experience designing and reviewing system architectures in a technical leadership capacity
- Solid understanding of data engineering concepts (SQL, ETL, cloud-based pipelines)
Preferred Qualifications
- Knowledge of model interpretability and fairness frameworks
- Experience with tools for bias detection and explainability in ML systems
- Hands-on experience with distributed data processing (e.g., Spark, pandas UDF)
- Background in financial systems, risk modeling, or regulated environments
- Familiarity with security and compliance standards in financial services
- Technology Environment
- Cloud & Infrastructure: Public cloud services (compute, storage, serverless)
- Data &; ML: Python, SQL, workflow orchestration toolsMLOps & DevOps: Infrastructure as code, CI/CD pipelines, version control systems
- Collaboration: Modern team communication and documentation platforms
- Language
Requirements
- Professional proficiency in Japanese (business level or above)
- Working proficiency in English (ability to read/write technical documentation)
