Principal Systems Software Engineer, Semiconductor Systems Inspection

NVIDIA•Santa Clara, CA

9d•$272,000 - $431,250

About The Position

NVIDIA is seeking a hands-on architect to define, build, evaluate, and demonstrate AI software for semiconductor inspection and manufacturing. The team develops models, adaptation and evaluation workflows, and deployable inference capabilities that fabs, equipment manufacturers, and software providers can integrate into their own tools. The work addresses limited and proprietary labeled data, distribution shifts across tools and fabs, tight production budgets, and deployment inside air-gapped facilities. The role spans computer vision, time-series modeling, multimodal AI, anomaly detection, model adaptation, evaluation, and production inference. The goal is to deliver technology that a fab or equipment vendor can integrate, operate, and trust in production—not merely demonstrate internally. The engineer will work directly with manufacturing partners and collaborate across research, software, process, metrology, inspection, and hardware teams.

Requirements

MS or PhD in Computer Science, Electrical Engineering, Computer Engineering, or a related technical field, or equivalent experience.
12+ years of professional software or algorithm engineering, including demonstrated technical ownership and architectural direction. Directly relevant PhD research may count toward this total, but research alone cannot satisfy the domain and production-delivery requirements.
8+ years in semiconductor manufacturing, including inspection, metrology, defect review, process control, yield engineering, or fab-data systems.
Hands-on delivery of software that shipped in semiconductor equipment, operated in a fab, or was adopted by a manufacturing customer.
4+ years of hands-on deep learning, machine learning, computer vision, or applied AI, including work within the last 24 months.
Strong Python skills and production experience with PyTorch or TensorFlow.
Track record of taking ambiguous technical problems through architecture, implementation, evaluation, and validated results.
Working GPU literacy, including ability to reason about inference performance and evaluate hardware and software bottlenecks.
Direct engagement with fabs, manufacturing teams, or equipment vendors on technical requirements.
Strong analytical, communication, and cross-functional leadership skills.

Nice To Haves

Experience adapting large pretrained vision, multimodal, or world foundation models to specialized industrial domains.
Background in anomaly detection and generation, time-series modeling, FDC, virtual metrology, or advanced-packaging inspection.
Experience designing evaluation frameworks for sparse proprietary data and shifting production distributions.
Expertise in synthetic data, self-supervised or few-shot learning, domain adaptation, model compression, and inference optimization using NVIDIA technologies.
Experience building SDKs, platform capabilities, or agentic workflows adopted by external engineering teams.

Responsibilities

Define and prototype AI system architectures for optical and e-beam inspection, wafer and mask inspection, metrology, defect review, equipment signals, and process data.
Advance world foundation model capabilities for semiconductor manufacturing, including vision, time-series and multimodal representation learning, model adaptation, domain transfer, and data-scarce defect understanding.
Develop workflows for defect detection, classification, localization, segmentation, nuisance filtering, ADC, ADR, and equipment time-series anomaly detection.
Establish evaluation methods to measure model quality and robustness under tool-to-tool and fab-to-fab domain shift, diagnose failures, and define production-readiness thresholds.
Design fab-local evaluation and adaptation workflows that preserve customer data sovereignty and produce comparable evidence across partner deployments.
Connect inspection and metrology results with equipment signals, process history, wafer genealogy, SPC, and yield outcomes to support manufacturing-wide reasoning and root-cause analysis.
Design agentic workflows for air-gapped fabs that connect data triage, model inference, review assistance, root-cause analysis, human approval, and secure deployment.
Optimize and package inference capabilities to meet production latency, memory, throughput, reliability, and security budgets, enabling partner deployment without NVIDIA operating partner systems.
Work directly with fabs and equipment manufacturers to move capabilities from prototype to partner-operated production.
Collaborate across research, software, process, metrology, inspection, and hardware teams to define platform capabilities and the technical roadmap for semiconductor AI.