h9 NgdE VHCexphmY - Remote — United States

About the Company

Prosum is recruiting for a client seeking an AI Data Scientist to develop advanced artificial intelligence solutions for industrial inspection and defect detection. The position focuses on applying machine learning, computer vision, and mathematical modeling to improve automated inspection technologies.

About the Role

The AI Data Scientist will develop and optimize Automated Defect Classification (ADC) and Die-to-Database (DB) comparison algorithms. The work involves turning advanced mathematical models into practical industrial applications that can identify defects and improve inspection accuracy.

You will work with deep learning, computer vision, image processing, and statistical modeling techniques to detect micro-scale defects, compare optical images against CAD database layouts, and improve model performance using real-world inspection data.

The position involves close collaboration with software engineers, field application teams, and domain experts to develop, validate, and deploy production-ready AI models.

Work Arrangement: Hybrid, with four days per week on-site and one day remote.

Work Locations: North Phoenix, Arizona, or Hillsboro, Oregon.

Key Responsibilities

• Design, train, and optimize deep learning and computer vision models for automated defect detection and classification.

• Develop object detection and image classification algorithms to identify micro-scale defects, including particles, scratches, and pattern anomalies.

• Build high-precision image registration, geometric alignment, and image-matching algorithms to compare captured optical images with CAD database layouts.

• Improve image alignment accuracy and reliability under varying imaging conditions.

• Apply statistical modeling, signal processing, denoising, and feature extraction techniques to improve image quality and inspection accuracy.

• Analyze large and noisy datasets to improve signal-to-noise ratios and model performance.

• Evaluate model performance using real-world inspection data and investigate challenging edge cases.

• Improve model accuracy through iterative training, active learning, and few-shot learning techniques.

• Collaborate with software engineers, field application teams, and domain experts to deploy and validate production-ready AI models.

• Document algorithms, experiments, and model performance results.

• Participate in code reviews, technical discussions, and knowledge-sharing activities.

• Support data preparation, annotation, validation, and quality assurance.

• Keep up with developments in computer vision, machine learning, and artificial intelligence.

• Work with cross-functional teams to support product development and model deployment.

• Perform other duties as assigned.

Required Qualifications

The original job listing does not specify formal education, years of experience, or mandatory technical qualifications. Candidates should review the role's technical responsibilities and relevant AI, machine learning, computer vision, and image-processing experience when assessing their suitability.

Preferred Qualifications / Requirements

The source listing does not identify a separate preferred qualifications section.

Relevant technical areas highlighted in the role include:

• Deep learning and computer vision.

• Object detection and image classification.

• Image registration, geometric alignment, and image matching.

• CAD database comparison.

• Statistical modeling and signal processing.

• Denoising and feature extraction.

• Active learning and few-shot learning.

These are technical areas involved in the position's responsibilities, not separately stated mandatory requirements.

Salary & Benefits

The advertised annual salary range is $115,000–$140,000.

The original listing does not provide details about additional benefits or bonus arrangements.

Work Arrangement & Employment Type

• Employment Type: Full-time, direct hire.

• Work Locations: North Phoenix, AZ, or Hillsboro, OR.

• Work Arrangement: Hybrid, with four days on-site and one day remote per week.