Senior Machine Learning Engineer

About the role

Amartha, a technology-driven financial ecosystem serving micro-businesses across Indonesia, is hiring a Senior Machine Learning Engineer in Shenzhen. The role covers designing, developing and productionizing ML models for credit scoring, underwriting, fraud detection, collections and portfolio risk management, and collaborating with data and software engineers to integrate them into scalable systems.

Key responsibilities

• Design, develop and productionize ML models for credit scoring, underwriting, fraud detection, collections and portfolio risk management.
• Build features from customer, transaction, repayment, behavioural and alternative data.
• Select and evaluate algorithms, with emphasis on explainable, high-performing models such as XGBoost.
• Define offline and online evaluation metrics aligned with lending outcomes and business objectives.
• Address class imbalance, data leakage, bias, model stability and changing customer behaviour.
• Build reliable training, validation, deployment, monitoring and retraining pipelines.
• Monitor model performance, calibration, drift, fairness and operational impact in production.
• Produce clear model documentation and explain decisions to risk, product, engineering, compliance and business stakeholders.
• Collaborate with data and software engineers to integrate models into scalable production systems.
• Conduct experiments and turn model improvements into measurable business outcomes.
• Explore practical LLM and agentic-AI applications such as document processing, underwriting assistance, investigation workflows and internal productivity tools.

Required profile

• Bachelor's or Master's degree in Computer Science, Machine Learning, Statistics, Mathematics or a related field, or equivalent practical experience.
• 5+ years of ML engineering or data scientist experience.
• Experience deploying, monitoring and maintaining ML models in production.
• Knowledge of credit risk, underwriting, fraud detection, customer scoring or related financial-services use cases; fintech, digital lending or microfinance experience strongly preferred.

Required skills

• Strong foundation in traditional machine learning: classification, regression, feature engineering, model evaluation and imbalanced-data handling.
• Hands-on experience with XGBoost, LightGBM, random forests and logistic regression.
• Proficiency in Python, SQL and ML libraries such as scikit-learn, XGBoost, pandas and NumPy.
• Understanding of model explainability, drift detection, experiment tracking and reproducible ML workflows.
• Familiarity with cloud platforms, containers, APIs and data pipelines; exposure to LLMs, RAG, prompt engineering or agentic programming is a plus.