This position aims to reinforce NVIDIA's semiconductor inspection roadmap by expanding operational capacity in a strategically meaningful area. The immediate focus centers on developing concrete AI products-models, adaptation workflows, and inference pipelines-building on the robust technical foundation already set with semiconductor customers and partners. This role balances the use of innovative methods with the delivery of practical systems that operate within tight deployment budgets in inspection environments.
NVIDIA is looking for a Sr. Software Engineer specializing in systems inspection. The role involves developing the next generation of AI products for semiconductor analysis in Santa Clara. This position will concentrate on redefining promising technical approaches into production-ready models and inference pipelines for key semiconductor manufacturing projects. The work focuses on computer vision, multimodal AI, anomaly detection, model compression, and deployment optimization. The team is currently developing innovative anomaly generation and inspection workflows for semiconductors. These workflows face challenges like limited data, domain shifts, and tight deployment requirements in fabrication facilities. This role is aimed at speeding up roadmap progress and turning research momentum into deployable AI products within a small, high-impact core team and consistently advancing model quality, robustness, and production readiness for challenging industrial inspection scenarios.
What you'll be doing:- Define and prototype AI system architectures for semiconductor defect inspection across optical inspection, e-beam inspection, wafer and mask inspection, metrology, and defect review workflows.
- Advance WFM capabilities for semiconductor inspection, including multimodal representation learning, model adaptation, domain transfer, and data-scarce defect understanding.
- Work with our partners to integrate and enhance existing computer vision and multimodal inspection workflows for defect detection, classification, localization, segmentation, nuisance filtering, ADC, and ADR.
- Design agentic inspection flows for air-gapped fab environments, connecting data triage, model inference, review assistance, root-cause analysis, human approval, and secure deployment constraints.
- Use semiconductor metrology, inspection, review, and process context, including CD, LER, LWR, overlay, wafer maps, defect maps, SPC signals, and yield signals, to improve model quality and fab decision support.
- Work with our partners to address noisy, limited, and shifting fab data, including tool-to-tool calibration, domain-shift mitigation, synthetic defect generation, noise simulation, and augmentation.
- Convert research into customer-ready semiconductor inspection products with clear evaluation, failure analysis, monitoring, optimization, and production deployment paths.
- Partner with research, software, process, metrology, inspection, review, and hardware teams to define roadmap priorities for next-generation semiconductor AI inspection systems.
What we need to see:- MS, or PhD in Computer Science, Electrical Engineering, Computer Engineering, or a related technical field, or equivalent experience.
- 3+ years of proven experience in deep learning, machine learning, computer vision, or applied AI.
- Strong programming skills in Python and experience with modern deep learning frameworks such as PyTorch or TensorFlow.
- Experience developing or applying foundational world models in computer vision for classification, detection, segmentation, anomaly detection, or multimodal understanding.
- Familiarity with self-supervised, few-shot, weakly supervised, unsupervised, or domain adaptation approaches relevant to inspection problems.
- Strong analytical, communication, and cross-functional collaboration skills.
Ways to stand out from the crowd:- Experience with semiconductor inspection, industrial visual inspection, manufacturing AI, metrology, or defect review workflows.
- Experience with knowledge distillation, model compression, quantization, pruning, or deployment optimization for edge or production environments.
- Background in anomaly detection or anomaly generation, especially in domains with unusual labels and shifting visual distributions.
- Familiarity with NVIDIA software and deployment tools such as TensorRT, CUDA, cuDNN, Triton, DeepStream, TAO Toolkit, or RAPIDS.
- Experience building end-to-end pipelines that span data curation, training, evaluation, export, and inference in production settings.
Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 152,000 USD - 241,500 USD.
You will also be eligible for equity and benefits.
Applications for this job will be accepted at least until June 20, 2026.
This posting is for an existing vacancy.
NVIDIA uses AI tools in its recruiting processes.
