Overview In this role, you will help build next-generation foundation models for unified, cross-scenario recommendations. You will advance event-sequence-driven generative paradigms and integrate multimodal understanding. You'll apply LLMs across retrieval, ranking, and re-ranking, while optimizing training, inference, and system design. You'll collaborate with a research-leaning, open environment to push the boundaries of intelligent recommendations, shaping a scalable, adaptable recommender platform.
Responsibilities- Build and optimize cross-scenario shared Foundation Models for unified modeling and efficient inference
- Advance event-sequence-driven generative recommendation, integrating multimodal understanding and generative capabilities
- Apply LLM technologies across retrieval, ranking, and re-ranking; contribute to model training, inference optimization, and system co-design
- Explore integration of LLMs/VLMs with recommendation systems to develop adaptive intelligent recommenders
- Research end-to-end generative recommendation and system optimization methods balancing efficiency and user experience
Key requirements- MS/PhD in Computer Science or related field or equivalent industrial research experience
- Strong theoretical foundation in ML, DL, or information retrieval
- Proficiency in Python and familiarity with PyTorch
- Strong passion for intelligent recommendation systems and a self-driven research mindset
- self-motivated research mindset
- strong problem-solving and analytical skills
- collaborative and open to sharing hypotheses and validations
- Python
- PyTorch
- machine learning fundamentals
