Data Scientist
irco
Job Description
- Own the end-to-end MLOps lifecycle – model packaging, versioning, cloud deployment, monitoring, and automated retraining pipelines on GCP using Vertex AI, MLflow, or Kubeflow.
- Design and maintain CI/CD pipelines for ML models, ensuring reliable, repeatable deployments with full model registry traceability from training data through to production artifacts.
- Define and enforce data quality governance standards across all ML feature pipelines and training datasets – including schema contracts, null checks, range validation, and detection of training-serving skew.
- Validate model outputs and analytical findings for statistical soundness and insights validation – reviewing for data leakage, biased evaluations, distributional assumptions, and reproducibility before results reach stakeholders.
- Set up model monitoring to track prediction drift, data drift, and performance degradation in production, and trigger automated retraining workflows when thresholds are breached.
- Work with large-scale IoT sensor datasets from industrial equipment such as air compressors and rotating machinery to build scalable, production-grade time-series and fault-detection pipelines.
- Collaborate with data engineers, domain experts, and product managers to translate requirements into scalable data science solutions, and clearly communicate model performance and business impact to technical and non-technical stakeholders. Actively use Gen AI coding assistants to accelerate development, generate boilerplate, write unit tests, and review code quality.