计算机视觉及多模态算法实习生（J106105）
Core
Develop computer vision and document processing algorithms, including OCR, document structuring, and multimodal large model training/fine-tuning for business applications.
Role type
Computer vision and multimodal algorithm intern
Builds
Production-ready multimodal models and document analysis features
Domain
Computer Vision, Natural Language Processing, Multimodal AI
Deliverable
production ML models
Required skills
Computer vision, OCR, document structuring, multimodal large models, PyTorch, PaddlePaddle, deep learning frameworks
Preferred skills
Top-tier conference papers, Kaggle/AIStudio/Tianchi competition experience, unsupervised/self-supervised learning
Technologies
OpenCV, PyTorch, PaddlePaddle
Responsibilities
Research and develop OCR and document analysis algorithms; Train and fine-tune multimodal large models; Deploy models in real-world business scenarios; Evaluate and optimize model performance; Collaborate on technical challenges
Seniority
Intern
