Define and drive the mid- to long-term technology roadmap for computer vision and AI, aligned with semiconductor equipment product lines (inspection, metrology, lithography, packaging, etc.).
Track industry standards (e.g., SEMI), advanced process nodes (e.g., 3nm/5nm), and emerging CV/AI technologies (few-shot learning, anomaly detection, foundation model fine-tuning), and lead technology pre-research initiatives.
2. Algorithm Development & Optimization (Core)
Lead the development of core computer vision algorithms for semiconductor applications.
Optimize algorithm performance in terms of speed, accuracy, and robustness, minimizing false positives/negatives in high-resolution, high-noise, and high-throughput production environments.
Architect and industrialize hybrid solutions combining classical image processing (OpenCV, Halcon) and deep learning models (CNNs, Transformers).
3. Team & Project Management
Build, lead, and develop a high-performing computer vision team, including talent pipeline development and technical capability building.
Manage the full project lifecycle: requirement analysis, task decomposition, timeline control, quality assurance, and release management.
Oversee code reviews, technical documentation (algorithm design, APIs, test reports), and promote knowledge standardization and reuse.
4. Cross-functional Collaboration
Collaborate with optics, hardware, and electrical teams on system-level design, including camera, lighting, sensor selection, and imaging system calibration.
Work closely with system software and embedded teams on algorithm integration and deployment optimization (GPU acceleration, CUDA, multi-threading), considering memory and power constraints.
Interface with manufacturing, quality, and customers for on-site issue resolution, data feedback loops, and algorithm customization.
5. Innovation & Intellectual Property
Lead key technology breakthroughs and drive the development of patents, software copyrights, and technical publications.
Establish end-to-end data pipelines covering dataset construction, training, evaluation, and deployment.
