Job Description

Responsible for developing intelligent solutions based on multimodal large models, promoting the application of AI technology in game products.

1. Multimodal AI Agent Development: Design and implement intelligent Agent systems based on vision-language models to support automated analysis and interaction in game scenarios.

2. Model Optimization and Deployment: Fine-tune, quantize, and engineer optimize multimodal large models to ensure efficient operation in production environments.

3. Visual Algorithm Research and Development: Develop visual algorithm modules such as object detection, image segmentation, and OCR for game scenarios.

4. Technological Innovation: Track cutting-edge AI technologies and explore innovative applications of LLM, AIGC, and other technologies in the gaming field.

Job Requirements

• Solid foundation in deep learning, familiar with Transformer architecture and its variants.
• Practical application experience with multimodal models (e.g., CLIP, BLIP, LLaVA).
• Proficient in object detection (YOLO, DETR, etc.) and image segmentation algorithms.
• Proficient in Python, familiar with PyTorch/TensorFlow, with good engineering implementation skills.
• 3+ years of relevant work experience.