Principal Engineer (Advance Data Analytics), Heterogeneous Integration Group(HIG), High Bandwidth Memory (HBM) Product Engineering

Originally posted 23 September 2026 by the employer.

This role focuses on building predictive models for High Bandwidth Memory (HBM) screening, dispositioning, and testing in high-volume manufacturing.

About the role

This role involves leading a team of engineers to build predictive models for High Bandwidth Memory (HBM) screening, dispositioning, and testing in high-volume manufacturing. You will work on intelligent systems to improve engineering productivity, strengthen technical decision-making, and unlock insights from complex manufacturing, validation, and engineering workflows.

What you'll do

• Lead a team of engineers, set technical direction, review modeling work, and manage workload.
• Own the design, training, validation, and productization of models that predict yield loss, defect escapes, and reliability risk.
• Build and operate the deployment path for models, including real-time and batch inference, monitoring, and drift detection.
• Develop and productionize machine learning and deep learning models for classification, regression, and anomaly detection.
• Develop scalable data pipelines and analytical workflows to ingest, clean, transform, and analyze large datasets.
• Apply GenAI and agentic systems to failure triage, root-cause analysis, and analysis automation.

What you'll need

• Bachelor’s or Master’s degree in Electrical Engineering, Computer Science, Data Science, Statistics, Artificial Intelligence, or a related field.
• Minimum 5 years of hands-on machine learning, data science, or predictive-modeling experience.
• At least 2 years leading a team of engineers or data scientists as a people manager or formal technical lead.
• Demonstrated depth in gradient-boosted tree methods such as XGBoost, LightGBM, or CatBoost.
• Practical experience with time-series and temporal modeling.
• Track record of models running in production, including deployment, performance monitoring, and retraining.
• Ability to define and defend business-level acceptance criteria for a model.
• Strong programming proficiency in Python and SQL.
• Strong technical foundation in data analytics and visualization, including tools and libraries such as pandas, scikit-learn, matplotlib, plotly.
• Experience analyzing large, complex, and heterogeneous datasets with techniques for data cleansing, outlier handling, and missing-data treatment.
• Cloud experience with GCP, AWS, or Azure, including deploying ML pipelines in production.
• Strong analytical, problem-solving, and software development skills.
• Strong communication skills.

Nice to have

• Experience with semiconductor manufacturing, test, or yield data.
• Direct experience in yield prediction or yield improvement modeling, die or wafer disposition, or reliability prediction.
• Experience with memory products, advanced packaging, 3D stacking, or heterogeneous integration.
• Deep understanding of semiconductor-specific AI/ML applications.
• Experience using enterprise data platforms such as BigQuery, Snowflake, MSSQL, Oracle, or Redshift.
• MLOps tooling and practice, including experiment tracking, feature stores, model registries, pipeline orchestration, and containerized deployment.
• Experience with Kubernetes or similar production infrastructure and deployment frameworks.
• Experience with anomaly detection, survival analysis, causal inference, uncertainty quantification, or Bayesian methods.
• Experience designing scalable, enterprise-grade AI/ML systems.
• Experience building agentic systems or AI solutions for semiconductor manufacturing, product engineering, validation, yield improvement, reliability, or failure analysis.
• Familiarity with agentic AI frameworks such as LangGraph, Google ADK, or AutoGen, evaluation tools such as AgentEval, and modern AI coding tools such as Claude Code, Cursor, Cline, Windsurf, or Gemini CLI.
• Experience working in cross-functional environments spanning engineering, manufacturing, data science, and IT.