Algorithm Engineer - Deep Learning

Overview In this role you will push the boundaries of Deep Learning and GenAI for image processing and defect detection in semiconductor process control. You join BBP, a division focused on high-sensitivity optical inspection for inline defect monitoring. You will translate state-of-the-art DL/GenAI models into practical, scalable solutions that improve throughput and deployment efficiency. This is a hands-on, end-to-end role with impact on product features and performance in a fast-paced, collaborative environment. Compensation / Benefitsmedical, dental, vision401(K) with company matchingemployee stock purchase program (ESPP)tuition reimbursementdevelopment and career growth opportunitieswellness benefits and EAP ResponsibilitiesUnderstand state-of-the-art deep learning and GenAI models and map them to domain problemsDesign DL/GenAI models to meet domain requirements and optimize architecturesImplement modeling prototypes and conduct rigorous analysesTrain and tune models on domain datasets and evaluate against defined metricsAnalyze performance bottlenecks and optimize throughput and cost (mixed-precision, FP16/FP8)Collaborate with peers and present results in technical settings Key requirementsPh.D. in Electrical Engineering, Computer Science, or related quantitative fieldExperience applying deep learning/GenAI to real-world problems with impactful resultsExpertise in Vision Foundation Models (VFM) or Vision Language Models (VLM) in CV, image processing, robotics, or NLPDL model optimization/distillation for mixed precision (FP16/FP8) to boost throughput and deployment efficiencyExperience with GenAI coding tools, vibe coding, or vibe engineeringProficiency in Python and at least one of C++, Java, Rust, GoProficiency in at least one DL framework (PyTorch, TensorFlow, JAX)Publications, patents, or impactful open-source work demonstrating deep learning expertiseTravel up to 10%Experience in semiconductor process control is a plusCollaborative mindsetStrong communication of complex ideasIndependent and self-drivenDeep learningVision Foundation Models (VFM)Vision Language Models (VLM)