We are looking for an experienced Data Scientist to support the development and continuous enhancement of a large-scale data and machine learning ecosystem used in next-generation automotive solutions.

The project involves building Spark-based data pipelines, improving data quality processes, implementing model evaluation workflows, and developing robust monitoring for ML models running in production.

The role includes both initial development of the platform components and long-term maintenance and optimization.

Initial Development:

• Develop Spark jobs for data ingestion and feature engineering
• Implement data quality monitoring (metrics, dashboards, alerting)
• Build logic for model evaluation and automated deployment decisions
• Develop model monitoring with visualized KPIs and technical metrics

Further Development / Maintenance:

• Continuously extend data pipelines and feature engineering workflows
• Enhance data quality metrics and monitoring coverage
• Expand model monitoring logic and dashboards
• Troubleshoot and fix code issues, including edge cases
• Experiment with new ML algorithms and additional data attributes
• Optimize performance and cost (algorithms, data structures, storage formats)
• Adjust training/deployment pipeline configurations (frequency, resources, etc.)

• 5+ years of experience
• Strong commercial experience with PySpark
• Excellent knowledge of Python
• Practical experience with GitHub
• Strong data analysis skills (Jupyter, Seaborn, exploratory analytics)
• Solid SQL knowledge
• Experience with Kubeflow or MLflow (MLOps frameworks for training, deployment & monitoring)
• Understanding of MLOps practices, including continuous training
• Experience with ML frameworks: scikit-learn, Pandas, Optuna
• Ability to create Grafana dashboards
• General knowledge of AWS services (S3, IAM, etc.)
• In-depth understanding of statistics and machine learning (missing data, outliers, model validation, algorithms)
• Fluent in Polish and good English

• Experience optimising data pipelines (Iceberg, Parquet, DynamoDB, etc.)
• Background in automotive or IoT data projects
• Experience with cost optimisation for ML systems
• Experience with large-scale model deployment pipelines