Intern - Engineer HVM PEE PHOTO

Location
Singapore, Fab10

Department
Fab10 High-Volume Manufacturing Photo Process and Equipment Engineering

Project Title
Automation of Non-Zero-Offset Control for High-Volume Manufacturing Photo Processes

Project Description

Non-Zero-Offset is a key photo-process control parameter used in overlay-performance management and yield protection. The current workflow includes data preparation, Non-Zero-Offset generation, engineering review, validation, and post-implementation monitoring. As the number of process vintages increases, a scalable analytical approach is needed to improve workflow consistency and efficiency.

The intern will undertake a structured project to develop and evaluate an automated, data-driven Non-Zero-Offset framework using Python, statistical analysis, machine learning, and visualization. The project will provide practical exposure to photo-process engineering, inline metrology, data modeling, workflow automation, and AI-enabled engineering analysis.

Objective of the Project

• Develop an understanding of the end-to-end Non-Zero-Offset workflow, including wafer selection, inline recipe criteria, metrology requirements, generation logic, and validation.
• Analyze historical Non-Zero-Offset, inline, and metrology data to identify patterns, risks, and improvement opportunities.
• Develop a Python-based analytical framework that improves the consistency and efficiency of Non-Zero-Offset analysis.
• Evaluate statistical, machine-learning, or approved AI-assisted methods for predicting Non-Zero-Offset behavior and potential risk conditions.

Project Scope

• Study the high-volume manufacturing photo-process flow and Non-Zero-Offset control methodology with relevant engineering subject matter experts.
• Prepare and analyze historical Non-Zero-Offset, inline, recipe, and metrology datasets using Python-based data pipelines.
• Develop and evaluate statistical or machine-learning approaches for identifying Non-Zero-Offset patterns, trigger conditions, and potential risk indicators.
• Design and prototype an automated logic flow, visualization, or dashboard that improves Non-Zero-Offset review and engineering decision-making.

Learning Opportunities

• Gain practical exposure to photo-process engineering, overlay control, inline metrology, and semiconductor manufacturing data.
• Learn Python-based data preparation, statistical analysis, modeling, visualization, and workflow-automation techniques.
• Develop familiarity with machine learning and approved AI-enabled tools for pattern identification, analytical interpretation, and technical documentation.
• Collaborate with photo-process owners and engineering subject matter experts to validate analytical results and translate findings into improvement recommendations.

Deliverables

• A cleaned, structured, and documented dataset containing relevant Non-Zero-Offset, inline, recipe, and metrology information.
• Reusable Python scripts for data preparation, Non-Zero-Offset analysis, modeling, and visualization.
• A validated statistical or machine-learning model for Non-Zero-Offset behavior, trigger conditions, or risk identification.
• An automation prototype and final technical presentation covering the methodology, results, limitations, recommendations, and future scaling opportunities.

Impact of the Project

• Reduce repetitive analytical steps within the selected Non-Zero-Offset workflow.
• Improve consistency and visibility in Non-Zero-Offset generation, review, and validation.
• Enable earlier identification of potential overlay-performance or yield-related risk conditions.
• Demonstrate a scalable analytical and automation framework for high-volume manufacturing photo processes.

Skillsets Required

• Proficiency in Python for data preparation, analysis, modeling, and visualization.
• Basic knowledge of statistics, machine learning, or applied data analytics.
• Strong analytical thinking, structured problem-solving, and ability to work with large datasets.
• Effective technical communication skills and familiarity with approved AI tools or AI-enabled analytical workflows.

Course of Interest

The ideal candidate should be pursuing a degree in Electrical Engineering, Electronic Engineering, Chemical Engineering, Mechanical Engineering, Industrial and Systems Engineering, Data Science, Computer Science, or a related field.

Duration

The ideal candidate should be able to commit to a full-time university internship period of five months, from January 2027 to May 2027.

Opportunities for Full-Time Employment

Interns may be considered for future internship or full-time employment opportunities based on business requirements, role availability, project outcomes, and the applicable recruitment process.