Senior Applied Scientist, Large Language Models

Team: Data & Technology

Location: Shanghai

Commitment: Full Time

Workplace Type: onsite

Key Responsibilities

Research and develop large language model capabilities for real-world applications.
Improve model performance in areas such as reasoning, long-context understanding, information extraction, retrieval-augmented generation, and domain adaptation.
Design and implement model post-training approaches, including supervised fine-tuning, preference optimization, knowledge distillation, and synthetic data generation.
Develop systematic evaluation methodologies covering accuracy, factuality, robustness, safety, latency, and cost.
Build scalable data preparation, model experimentation, and evaluation pipelines.
Analyse model failure cases and identify effective approaches for continuous improvement.
Explore emerging research and assess its practical value in production environments.
Work with engineering teams to deploy and optimize models and AI capabilities in production.
Collaborate with product managers and domain experts to translate business requirements into algorithmic solutions.
Contribute to technical standards, best practices, and the longer-term development of the AI technology roadmap.
Provide technical guidance and support to other algorithm engineers and researchers.

Qualifications

Master’s degree or PhD in Computer Science, Artificial Intelligence, Machine Learning, Natural Language Processing, or a related discipline, or equivalent practical experience.
Strong experience in machine learning, natural language processing, or applied AI.
Hands-on experience developing or adapting large language models.
Strong understanding of Transformer architectures, model training, fine-tuning, and inference.
Practical experience in at least two of the following areas:
LLM post-training and alignment
Model evaluation and benchmarking
Retrieval-augmented generation
Long-context modelling
Information extraction
Complex reasoning and planning
Model compression or inference optimization
Strong proficiency in Python and deep learning frameworks such as PyTorch.
Ability to independently define algorithmic problems, design experiments, analyse results, and deliver production-ready solutions.
Strong communication and cross-functional collaboration skills.

Preferred Qualifications

Experience developing AI solutions for enterprise, scientific, technical, or other knowledge-intensive applications.
Experience with distributed training, large-scale inference, or GPU optimization.
Experience building automated evaluation systems, data flywheels, or human-feedback pipelines.
Experience with multimodal models, AI agents, or tool-augmented language models.
Publications in reputable AI, machine learning, or NLP conferences, or meaningful open-source contributions.