Data Scientist

Role Overview

We are looking for an experienced Data Scientist with over five years of expertise to create machine learning solutions aimed at predicting failures, classifying faults, and analyzing equipment issues within semiconductor manufacturing and related systems. The role emphasizes working with time-series data, monitoring equipment health, and performing root-cause analysis by leveraging structured reliability approaches like fault tree analysis (FTA).

Key Duties and Responsibilities

• Design, develop, and implement machine learning models focused on predicting equipment failures and categorizing faults.
• Work extensively with time-series datasets derived from semiconductor tools, including sensor signals, logs, and process-related data.
• Employ advanced feature engineering techniques such as lag features, rolling windows, trend extraction, seasonality analysis, and event-based characteristics.
• Incorporate fault tree analysis principles to facilitate root-cause investigations and enhance model explainability.
• Collaborate effectively with process engineers, equipment specialists, and failure analysis teams to deliver actionable insights.
• Choose and justify suitable machine learning algorithms while evaluating their performance rigorously.
• Use metrics like precision, recall, F1-score, ROC-AUC, and early failure detection accuracy for model validation.
• Maintain thorough documentation on models, underlying assumptions, and outcomes for both technical and cross-departmental audiences.
• Mentor junior data scientists and contribute to establishing best practices within the team.

Required Experience and Skills

• Minimum five years as a Data Scientist or Machine Learning Engineer.
• Strong command of Python programming, including libraries such as Pandas, NumPy, and scikit-learn.
• Demonstrated expertise in modeling time-series data.
• Experience building predictive and classification models for complex datasets.
• Background in failure prediction, reliability analytics, and equipment health monitoring.
• Practical knowledge of fault tree analysis or structured root-cause analysis techniques.
• Expertise in feature engineering tailored to noisy, real-world industrial data.
• Ability to communicate complex technical results clearly to engineering teams.

Preferred Qualifications

• Experience in semiconductor manufacturing areas such as etching, deposition, lithography, inspection, or metrology.
• Familiarity with process-related data, tool logs, alarm systems, and sensor telemetry.
• Knowledge of survival analysis methods, remaining useful life estimation, or detecting anomalies.
• Understanding of model interpretability tools like SHAP and feature importance evaluations.
• Hands-on experience deploying machine learning models into production or factory environments.
• Background or familiarity with reliability engineering, systems engineering, or failure analysis.

Indicators of Success

• Development of dependable failure prediction models characterized by low false positives.
• Clear association between data-driven predictions and physical causes of failure.
• Observable improvements in equipment uptime, manufacturing yield, and maintenance scheduling.
• Effective collaboration with cross-disciplinary engineering teams.

Technical Environment

• Python and its scientific stack including Pandas, NumPy, and scikit-learn.
• Libraries specialized in time-series data analysis.
• Use of various machine learning frameworks.
• Visualization and reporting tools to communicate findings.