AI & Data Senior Scientist

岗位职责
1. Lead the selection, fine-tuning, domain adaptation, and evaluation system construction of biomedical LLMs, iterating and optimizing models for proteomics, post-translational modification (PTM), and mass spectrometry data scenarios.
2. Architect and develop bioinformatics analysis agent systems, including tool orchestration, RAG knowledge base construction, hallucination mitigation, and agent performance evaluation, enabling automated and interactive execution of proteomics bioinformatics analysis tasks.
3. Lead data governance, dataset construction, training, and application development for multi-source omics data (raw mass spectrometry data, protein databases, literature, clinical samples).
4. Conduct data science, statistical mining, and multi-omics integration algorithm research in conjunction with proteomics business; deliver algorithm proposals and technical documentation; drive algorithm engineering implementation.
5. Track cutting-edge technologies in AI LLMs and computational biology; conduct technology foresight research, delivering technical surveys and prototype validation; lead a small team in algorithmic problem-solving; coordinate with clients and bioinformatics teams for project delivery.
6. Consolidate technical achievements, supporting patents, publications, and internal technical reports.

岗位要求
1. Education:​ Ph.D. in Computer Science, Artificial Intelligence, Bioinformatics, Computational Biology, or related fields.
2. Experience:​ 1-3 years of R&D experience in LLMs/computational biology/bioinformatics algorithms; outstanding fresh Ph.D. graduates are also welcome.
3. Key Technical Skills:​ Proficient in LLM fine-tuning (LoRA/SFT), RAG, Agent frameworks (LangChain/LangGraph); skilled in Python and PyTorch; experienced in dataset construction, model evaluation, and statistical analysis; familiarity with proteomics/mass spectrometry data is preferred.
4. Others:​ Experience with biomedical LLMs, bioinformatics agents, or multi-omics data mining projects is preferred; familiarity with Docker, vLLM, and inference deployment is preferred.