Artificial Intelligence Engineer

Responsibilities:

Own end-to-end data processing pipelines for display system characterization data (sensor images, metrology measurements, yield data) across multiple product builds
Develop ML/AI algorithms to automatically identify and classify visual artifacts, display defects, and performance anomalies in sensor and camera data
Build automated analysis tools for disparity sensor performance evaluation, including SNR estimation, pattern detection accuracy, and ambient cross-talk assessment
Design and implement anomaly detection models to flag display performance regressions in manufacturing and integration test data
Create data visualization dashboards and reporting tools to communicate display quality metrics to cross-functional hardware teams
Develop image processing algorithms for waveguide characterization including uniformity analysis, efficiency mapping, and defect detection
Collaborate with optical, process, and integration engineers to translate hardware requirements into algorithmic solutions and validate model performance against ground truth
Maintain and improve data infrastructure (collection, storage, versioning, and access) supporting the team's ML and analytics workflows
Document methodologies and contribute to team knowledge base for reproducible analysis.

Minimum Qualifications:

M.S. or Ph.D. in Electrical Engineering, Computer Science, Optical Engineering, Applied Physics, or a related quantitative field
3+ years of experience in ML/AI algorithm development for image processing, signal processing, or sensor data analysis
Strong proficiency in Python and experience with ML frameworks (PyTorch
Experience with image processing and computer vision techniques (feature detection, segmentation, classification, pattern matching)
Demonstrated ability to build and maintain data processing pipelines for large-scale experimental or manufacturing data
Experience with statistical analysis, hypothesis testing, and experimental design
Strong problem-solving skills with ability to work through ambiguous, hardware-related technical challenges
Excellent communication skills ability to present data-driven findings to cross-functional engineering teams

Preferred Qualifications:

5+ years of relevant industry experience in optics, display systems, or semiconductor/hardware characterization
Experience with display metrology MTF, luminance uniformity, chromaticity, contrast measurements
Familiarity with optical system modeling and ray-tracing concepts (Zemax, Code V, or equivalent)
Experience with deep learning for defect detection or anomaly classification in manufacturing contexts
Knowledge of AR/VR display technologies waveguides, micro-LEDs, LCoS, holographic optical elements
Experience with sensor characterization SNR analysis, noise modeling, dynamic range assessment
Proficiency with data visualization tools (Plotly, Matplotlib, Tableau, or Unidash)
Experience with version control (Git), collaborative development environments, and CI/CD pipelines
Familiarity with Client s internal tools and data infrastructure is a plus

Must-Have HARD Skills:

Python + PyTorch for ML/AI algorithm development
Image processing / computer vision (feature detection, segmentation, classification, pattern matching)
Building & maintaining data processing pipelines for large-scale experimental/manufacturing data

Nice-to-have Skills:

Display metrology (MTF, luminance uniformity, chromaticity, contrast) and sensor characterization (SNR, noise modeling)
Deep learning for defect/anomaly detection in manufacturing
AR/VR display tech (waveguides, micro-LEDs, LCoS, HOEs) + viz tools (Plotly/Matplotlib/Tableau/Unidash)
Past MAANG experience is a nice to have

Years of Experience: 5+ preferred - definitely within the scope of requirements mentioned above
Degrees/Certifications Required: M.S. or Ph.D. in EE, CS, Optical Eng, Applied Physics, or related engineering field

How many rounds of interviews:

1 -2 rounds max
1st round - 55 mins
2nd round - 15-30 mins

Types of Interviews: Technical (ML/CV + signal/image processing), a coding/data exercise, and a presentation of past relevant work; behavioral for collaboration.