Applied AI Engineer

Advantest America; San Jose (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Advantest America is seeking an Applied AI Engineer (Software Engineer) to build intelligent systems that automate, optimize and validate PCB design workflows. You will work at the intersection of electronics engineering, EDA tools and AI to reduce design cycle time and improve quality.

The role involves working with large-scale datasets and developing ML and RL solutions, including graph-based methods and optimization algorithms, to enhance PCB design processes and outcomes.

Qualifications

• Experience building AI/ML models for PCB design workflows.
• Strong understanding of ML fundamentals and evaluation metrics.

Responsibilities

• Collect, clean and preprocess structured and unstructured data from multiple sources (EDA software etc.)
• Build working prototypes for AI-assisted PCB Design automation.
• Design, train and evaluate supervised, unsupervised and RL (reinforcement learning) machine learning models.
• Implement models such as regression, classification, clustering, time series, GNNs, reinforcement learning and optimization algorithms.
• Formulate automation task as an optimization/RL problem, including state representation, action space, reward design, constraints and evaluation criteria.
• Evaluate and implement different algorithms such as simulated annealing, greedy approach, graph-based, force-directed methods, constraint solving, ILP/CP-SAT or evolutionary algorithms.
• Develop RL or learning-guided methods using realistic EDA/PCB design data.
• Define quality metrics for evaluating layout or assignment solutions based on cost efficiency, design-rule violations, conflict minimization, density and engineering review effort.
• Create benchmark datasets and evaluation pipelines to compare generated output against baselines and engineer-reviewed layouts.
• Design data representations for components, nets, board regions, keep-out zones, mechanical boundaries, constraints and connectivity graphs.
• Build visual/debug tooling to inspect output, failure cases and quality metrics.
• Work with PCB/layout/domain experts to translate design rules and PCB Design practices (placement, routing etc.) into software constraints.
• Contribute to the path from research prototype to usable engineering workflow.
• Ensure AI solutions follow ethical, responsible and explainable AI practices.

Skills

reinforcement learning

graph neural networks

optimization algorithms

data preprocessing

EDA tools

Tools

EDA tools

CP-SAT solver

simulated annealing

Job description

Position Overview: We are seeking highly skilled Applied AI Engineer (Software Engineer) to build intelligent systems that automate, optimize and validate PCB design workflows.

The person will work at the intersection of electronics engineering, EDA tools and AI to significantly reduce design cycle time, improve quality and enable next generation autonomous PCB design capabilities.

The role involves working with large-scale datasets, reinforcement learning, optimization, algorithms, building predictive and deploying ML/AI solutions for complex PCB Design workflows.

What You Will Do

• Collect, clean and preprocess structured and unstructured data from multiple sources (EDA software etc.)
• Build working prototypes for AI-assisted PCB Design automation.
• Design, train and evaluate supervised, unsupervised and RL (reinforcement learning) machine learning models.
• Implement models such as regression, classification, clustering, time series, GNNs, reinforcement learning and optimization algorithms.
• Formulate automation task as an optimization/RL problem, including state representation, action space, reward design, constraints and evaluation criteria.
• Evaluate and implement different algorithms such as simulated annealing, greedy approach, graph-based, force-directed methods, constraint solving, ILP/CP-SAT or evolutionary algorithms.
• Develop RL or learning-guided methods using realistic EDA/PCB design data.
• Define quality metrics for evaluating layout or assignment solutions based on cost efficiency, design-rule violations, conflict minimization, density and engineering review effort.
• Create benchmark datasets and evaluation pipelines to compare generated output against baselines and engineer-reviewed layouts.
• Design data representations for components, nets, board regions, keep-out zones, mechanical boundaries, constraints and connectivity graphs.
• Build visual/debug tooling to inspect output, failure cases and quality metrics.
• Work with PCB/layout/domain experts to translate design rules and PCB Design practices (placement, routing etc.) into software constraints.
• Contribute to the path from research prototype to usable engineering workflow.
• Ensure AI solutions follow ethical, responsible and explainable AI practices.