美商圖策生技有限公司台灣分公司

We seek an experienced Senior Data Scientist to join our innovative drug discovery team. The successful candidate will be responsible for developing advanced algorithms for antibody, small molecule, and nucleotide drugs using deep learning and reinforcement learning techniques. You will lead the design and preparation of datasets and databases, including 3D protein generation, and apply computational approaches such as AI-driven Drug Design (AIDD) and force field simulation. Additionally, you will enhance the performance of the deep learning model using edge computing. As a senior team member, you will collaborate with cross-functional teams to integrate AI technologies into our drug discovery processes and mentor junior data scientists, providing technical guidance and support. What You'll Do: • Develop the algorithms for structure-based drug generation • Take advantage of different algorithms/model architectures (diffusion-based, language-based, etc.) to suit the diverse purposes in drug development. • Support the integration of drug-related databases (e.g. PDB, ChEMBL, etc.) and establishment of the training-ready datasets, including pre-processing of data characterization and featurization for experimental and other data. Minimum Qualifications: • Proficiency in Python • 3+ years in solving biological/clinical problems with DL & RL • 1+ year in protein structure problem-solving with DL • 1+ year developing DL/RL algorithms • Strong understanding of statistics and probability • Familiarity with machine learning, deep learning, and PyTorch • Experience with version control (e.g., Git) • Proficient in software development and writing production-quality code Nice to Have: • Knowledge of SE(3) equivariant algorithms and 3D rotation • Experience in off-line policy Deep RL (e.g., SAC, TD3) • Familiarity with Linux OS and HPC • Debugging and performance profiling skills for Python/C++ • Strong biochemistry background • Experience in protein/nucleotide feature encoding

We seek an experienced Data Engineer to join our innovative drug discovery team. The candidate will build tens of thousands of chemical and experimental data into a data pool. Will support the design and preparation of datasets and databases, including data collection, cleaning, normalization, and export. Additionally, you will create an automated data update pipeline and develop visualization tools. As a team member, you will work with cross-functional teams to integrate AI technologies into our drug discovery processes and mentor engineers to provide technical guidance and support. What You'll Do : o Build large-scale drug and experiment-related databases, extract, transform, clean, normalize, and integrate data from different databases. o Support the team leader in creating training datasets for model development. o Collaborate with other teams to develop database visualization tools and create related APIs for AI engineers. o Perform statistical analysis of outliers and clustering using a background in medicinal chemistry o Establish the pipelines of automated data updates. Minimum Qualifications : o Must have a background in biology or chemistry. o Proficiency in Python o Experience in handling and organizing related databases such as Chembl/PubChem, etc. o Experience using FastAPI or other RESTful APIs. o Experience building PyTorch datasets/dataloaders o Familiarity with various chemical notations and the ability to understand different biochemical experiment theories and methods o Experience with multithreading and handling data on a billion-scale. o Have a strong ability of problem-solving and ability to collaborate and work as part of a team Nice To Have : o Experience handling different types of drug data(Small molecular, Antibody …). o Familiarity with 3D structural data o Experience building DL/RL-related models or algorithms o Good debugging and performance profiling skills for Python