Machine Learning Scientist

• Design and develop high-impact machine learning models for applied AI use cases, including demand and load forecasting, delay and disruption prediction, anomaly detection across operational and sensor data, and intelligent document and record classification.
• Select the most appropriate modelling approach for each problem—such as gradient boosting, statistical modelling, or deep learning—and clearly articulate the trade-offs, business value, and performance improvement over simpler baselines during solution reviews.
• Partner with team leads and business stakeholders to define evaluation datasets, success metrics, and target accuracy before development begins, then communicate results transparently against agreed objectives.
• Create robust evaluation sets with graded reference answers to support reliable model assessment, benchmarking, and continuous improvement.
• Collaborate with engineering teams to build scalable data pipelines from client operational data to model-ready inputs, including validation rules that proactively identify and reject poor-quality data before training or inference.
• Take models from experimentation into production with appropriate monitoring for input drift, output quality, and clearly defined retraining triggers to ensure sustained performance in real- world environments.
• Deliver maintainable, production-ready code supported by automated tests, clear documentation, and well-tracked experiments.
• Validate model outputs with business stakeholders to ensure they are accurate, explainable, and aligned with operational decision-making needs.
• Conduct structured error analysis after each delivery iteration, identify root causes, and recommend practical improvements to strengthen model performance and business outcomes.
• Document data lineage, model behaviour, assumptions, limitations, and key decisions to support governance, maintainability, and stakeholder confidence.

• Master’s or PhD degree in Computer Science, Statistics, Applied Mathematics, Machine Learning, Artificial Intelligence, or a closely related quantitative field.
• At least 3 years of hands-on experience developing and deploying machine learning models that are used in real business or operational environments.
• Strong proficiency in Python and modern machine learning frameworks such as PyTorch, TensorFlow, Scikit-learn, XGBoost, or equivalent technologies.
• Solid understanding of classical machine learning, statistical modelling, regression, gradient boosting, deep learning, and the ability to choose the right level of model complexity for each problem.
• Proven experience working with complex operational data, including feature engineering, missing-value handling, sensor noise, inconsistent formats, and data quality issues across multiple source systems.
• Experience building or maintaining model evaluation pipelines, including test-set design, metric selection, benchmarking, regression testing, and performance tracking across model versions.
• Ability to explain model behaviour, assumptions, limitations, and business implications clearly to non-technical stakeholders and operational teams.
• Strong software engineering discipline, including clean coding practices, version control, automated testing, documentation, and reproducible experimentation.
• Proficient English communication skills, with the ability to engage effectively in business, technical, and cross-functional discussions.

Preferred:

• Experience with time-series, forecasting, sensor data, anomaly detection, or predictive analytics in real-world systems is highly desirable.
• Experience developing machine learning solutions for large-scale time-series, sensor, or operational datasets, including forecasting, anomaly detection, predictive maintenance, or reliability analytics.
• Exposure to aviation, aerospace, mobility, or mission-critical operational domains, such as flight operations, ACARS/QAR data, MRO records, asset utilisation, or passenger demand forecasting.
• Hands-on experience with retrieval-augmented generation solutions, including retrieval quality tuning, chunking strategies, grounding evaluation, and answer-quality assessment.
• Practical MLOps experience in production environments, including model registries, experiment tracking, monitoring, automated retraining, and deployment governance.
• Familiarity with regional or domestic AI ecosystems and frameworks, such as MindSpo

• Direct interface with senior leadership in Singapore.
• A defined career development pathway within a global organization operating across aerospace, smart city, marine, and digital systems.
• Competitive total compensation package

Báo cáo cho

GM