Responsibilities
1. Design and build multimodal models for specific fields, including multimodal pre-training, instruction fine-tuning, human preference alignment, etc. 2. Use large-scale and multimodal LLM technology to solve technical problems related to the intelligence of various types of special effects materials 3. According to the business needs of CapCut, combine computer vision and deep learning technologies to produce, optimize and implement intelligent technology solutions 4. Participate in the research and development of Jianying's business needs and implement technical solutions to the business line.
Qualifications
1. Computer science, software engineering related majors, those with programming/mathematical modeling competition awards and top conference papers are preferred familiar with large language models and multimodal common algorithms, and proficient in Transformer and other algorithms 2. Strong hands-on practical ability, with multimodal large model pre-training/fine-tuning experience is preferred 3. Have excellent programming skills, familiar with Python/C++ programming, familiar with basic data structures and algorithms 4. Have experience in computer vision, image processing, and computer graphics 5. Have the ability to quickly learn new technologies, be able to understand cutting-edge papers in a relatively short period of time, and have independent thinking, problem analysis and problem solving capabilities 6. Strong sense of responsibility, proactive, love technology, and have good communication and teamwork skills.
