Job description
Responsibilities:
- Optimize the pipeline and model performance of large language models (LLMs) in retrieval-augmented generation (RAG) scenarios.
- Explore and develop advanced applications of LLMs in multi-modal tasks, function calling, and interactive search.
- Track the latest advancements in evaluation capabilities of large language models in RAG, multi-modal agents, and function calling, and build systematic evaluation capabilities.
- Optimize foundational abilities of LLMs (e.g., instruction following, reasoning and planning, long-text memory, and knowledge inheritance) and promote their practical implementation.
Qualifications:
- Bachelor’s degree or higher in Computer Science or related fields.
- Background in NLP/search/advertising/recommendation system development, familiar with modern NLP model architectures such as Transformers/BERT.
- Excellent coding skills, proficient in programming languages such as Python & Shell.
- Excellent communication and logical expression skills, a passion for exploratory learning, a good team collaboration attitude, and a strong sense of responsibility.
- Preferred: impactful work (e.g., publications in academic conferences, open-source contributions).