Advanced Analytics Specialist

Gravitas Recruitment Group Hong Kong

Posted 22 hours ago Permanent Competitive

Role Summary Support the development, maintenance and operationalisation of machine learning and advanced analytics solutions, with a primary focus on maintaining and enhancing existing analytical models, feature engineering pipelines and Python/PySpark-based codebases. Work closely with the Advanced Analytics team to improve code quality, end-to-end model reliability, maintainability and production readiness, solving complex business problems through advanced analytics, machine learning and AI. Job Description Model & Feature Maintenance

Maintain and enhance existing machine learning models, feature engineering logic and analytical pipelines using Python, PySpark and SQL.
Refactor existing code to improve readability, modularity, reusability, performance and maintainability.
Support model retraining, recalibration, feature updates and validation, ensuring consistent implementation across model training and scoring processes.

Advanced Analytics & Model Development

Support the development, validation and enhancement of predictive models and machine learning solutions, such as propensity, churn, customer lifetime value, recommendation and forecasting models.
Perform model evaluation, calibration, back-testing, feature selection and hyperparameter tuning.

Translate defined business requirements into practical and maintainable analytical deliverables.

MLOps & Productionisation

Support established machine learning lifecycle processes, including testing, deployment, monitoring and retraining.
Convert prototype code into reusable, maintainable and production-ready Python, PySpark and SQL components.
Maintain model pipelines, experiment records, model artifacts, dependencies and deployment configurations; support automated testing, version control, job scheduling and release management.
Collaborate with data engineering and technology teams, adhering to established MLOps, model governance, security and documentation standards.

Model Monitoring & Issue Resolution

Monitor model performance, data quality, feature availability, prediction distributions and model stability.

Investigate model deterioration, data and feature drift, scoring anomalies, code defects and pipeline failures, and support production issue resolution.

AI Use Case Support

Support selected AI use cases involving Generative AI, LLMs, RAG or AI agents through prototyping, testing and evaluation.

Qualifications and Experience

Bachelor’s degree in Data Science, Statistics, Mathematics, Computer Science, Engineering, Actuarial Science or another related quantitative discipline.
Minimum 5 years of relevant experience in data science, machine learning, advanced analytics, data and AI engineering, or a related quantitative field.
Proven experience deploying advanced analytics or machine learning solutions in a business environment.
Experience collaborating with analytics, business, IT and governance teams.

Skills/Knowledge

Strong hands-on programming skills in Python, PySpark and SQL, including debugging, refactoring and code optimisation.
Practical experience developing, maintaining, testing and enhancing machine learning models and analytical pipelines.
Practical experience with machine learning libraries such as Scikit-learn, XGBoost, LightGBM or equivalent.
Practical experience with supervised machine learning, predictive modelling and statistical analysis.
Practical experience in feature engineering, data preprocessing, model validation and performance analysis.
Familiarity with MLOps practices, including deployment pipelines, version control, automated testing, CI/CD and production support.
Exposure to Generative AI, LLMs, vector databases, RAG architectures or agentic workflows would be an advantage.