Job Responsibilities

1.Conduct research and optimization on content generation models (text/image/audio/3D model, etc.), solve existing algorithmic problems in generation quality, diversity, controllability and generation efficiency, ultimately improve user experience and production efficiency, and assist the productization and landing of algorithms.

2.Engage in algorithm training and optimization in the fields of image generation, multimodal large models, and few-shot learning.

3.Combined with internal product and business requirements, optimize the effect and experience of AI painting, AI text generation, AI video generation from aspects including prompt optimization, generative model R&D, adapter development, and performance acceleration; solve algorithm bottlenecks in business-oriented application and industrial landing scenarios.

4.Tackle industrial deployment challenges of multimodal generative models, and conduct in-depth research on model design and optimization.

Job Requirements

1.Full-time Bachelor’s degree (preferred), Master’s degree or above in Computer Science, Artificial Intelligence, Mathematics or related majors.

2.Solid theoretical foundation in computer vision or machine learning; candidates with top-tier conference/journal papers in relevant directions are preferred.

3.Proficient in the basic principles of machine learning and deep learning; familiar with mainstream AI generative model frameworks including GAN, VAE, VQGAN, Diffusion, etc.

4.Familiar with generative plugin models such as ControlNet, LoRA, Text Inversion.

5.Familiar with multimodal Transformer cross-modal representation models such as CLIP and ERNIE-ViL; practical experience in NLP, multimodal learning or AI generative models is preferred.

6.Strong learning ability, clear logical thinking, excellent communication skills and strong intellectual curiosity.

7.Good teamwork and cross-team communication capabilities.