
Machine Learning Engineer
- Jiangmen, Guangdong
- Permanent
- Full-time
- Translate business problems into ML tasks: predictive maintenance, image segmentation and classification, price quotation and forecasting, etc.
- Build data pipelines (PySpark, Synapse, Databricks) and feature engineering workflows.
- Train, fine-tune, and evaluate ML models (scikit-learn, XGBoost, PyTorch, TensorFlow) following experiment-tracking standards (MLflow).
- Containerize models; Deploy to AKS/edge devices via automated CI/CD pipelines (AML pipelines, Azure DevOps).
- Establish monitoring suite (Prometheus, Grafana, PromptFlow) for model, and data drift.
- Apply best-practice MLOps patterns: provenance, reproducibility, automated retraining, and rollback strategies.
- Co-create user stories with product owners, size tasks, and deliver incremental value in sprints.
- Produce clean, test-covered, well-documented code; participate in peer reviews.
- Conduct workshops and demos to upskill factory engineers & operators.
- 3 – 5 years hands-on experience in ML engineering or data science deploying models to production.
- Solid foundation in traditional ML, statistics, and experimentation (p-values, A/B, power analysis).
- Solid Python programming; experience with unit/integration testing frameworks (pytest).
- Practical knowledge of containerization (Docker) and at least basic Kubernetes concepts (pods, services, config-maps, secrets).
- Familiarity with Azure ML or comparable cloud ML services.
- Familiarity with Generative frameworks like LangChain, LlamaIndex etc to implement Agentic Flows
- Understanding CI/CD & IaC workflows (Git, GitHub Actions or Azure DevOps, Terraform/Bicep).
- Strong communication skills, curiosity to learn manufacturing processes, and bias for hands-on problem solving.