Machine Learning Data Scientist

We are seeking a Machine Learning Data Scientist to join our applied research team, where we develop real-time decision-making and recommendation systems for industrial production lines. You'll be responsible for developing robust and interpretable machine learning models that transform complex sensor data into actionable insights.

This role requires a strong mix of analytical thinking, practical ML experience, and the ability to work with real-world, noisy datasets. You'll collaborate closely with deep learning researchers, research engineers, and domain experts to design experiments, evaluate model performance, and bring impactful solutions to production.

Duties and Responsibilities

- Design, train, and evaluate machine learning models (e.g.,classification, regression, clustering, ranking) for real-time industrial applications.

- Engineer meaningful features from raw sensor and control data.

- Analyze large-scale datasets to uncover patterns, define KPIs, and guide algorithm development.

- Validate model assumptions, monitor performance over time, and detect model drift or anomalies.

- Work with data engineers to define data requirements and prepare clean, structured datasets.

- Collaborate with research engineers to operationalize ML models in production environments.

- Document findings and present insights to both technical and non-technical stakeholders.

Education and Experience

- B.Sc./M.Sc.in Computer Science, Statistics, Applied Mathematics, or related field.

- 3+ years of experience applying machine learning to real-world problems.

- Strong skills in Python and ML libraries such as scikit-learn, XGBoost, LightGBM, orCatBoost.

- Experience with feature engineering, model selection, hyperparameter tuning, and evaluation techniques.

- Solid understanding of statistics, data distributions, and uncertainty estimation.

- Hands-on experience working with messy, high-dimensional, or time-series data.

Other Skills

- Background in industrial, sensor-based, or time-series prediction problems.

- Familiarity with anomaly detection and root cause analysis.

- Experience working closely with MLOps or deploying models to production.

- Comfort working in fast-paced environments with evolving data and priorities.