Novartis Biomedical Research is seeking an experienced and highly motivated data and MLOps scientist/engineer/wizard to help us push the frontiers of data science and machine learning for Life Sciences and drug discovery. Within the Research Informatics division on Biomedical Research, you will take on a hands-on data scientist role at the intersection of science, RD and real-world impact. You will be a part of a truly unique organization with an inter-disciplinary team made up of accomplished scientists who are at the forefront of AI/ML in drug discovery. You will have the opportunity to shape the next wave of drug discovery, combining insights from a wealth of data modalities including various omics datasets (genomics, transcriptomics, proteomics), spatial omics technologies, compound structures, protein sequences and structures, compound activity, protein structures, measurements from cellular experiments and safety studies, histopathology, clinical imaging, and clinical readouts. This is a senior individual contributor role and will require demonstrated and current hands-on experience in Python, R, cli/shell scripting, diverse ML workflows, who can perform under minimum supervision in a highly collaborative environment. Key Responsibilities: Collaborate closely with data scientists and subject-matter experts to fulfill data and computational needs. Validate and ensure the accuracy and quality of data by cleaning, shaping, and sometimes analyzing, normalizing, and conforming it to existing models and vocabularies. Identify and rectify data inconsistencies and irregularities. Design data models and prepare data artifacts to effectively meet business needs. Promote culture of transparency and communication regarding data modifications, lineage, and definitions to all stakeholders Essential Requirements : Python and/or R, and any other scripting language. Experience with HPC, cloud (AWS) workflows, setup and deployments. Experience deploying ML models along with resolving package dependencies) from github, huggingface, etc, successfully. Experience with DevOps, MLOps Demonstrable data management expertise in relational, document, column and graph datastores. Experience building ETL processes in high-performance environments like Databricks, AWS, Snowflake. Experience with python ML frameworks: Pytorch, tensorflow is a plus. Experience with several bioinformatics tools for sequence matching, alignment, clustering is a plus. data imputation and visualisation methods. Strong background in extracting relevant data from diverse sources (excel sheets, powerpoint, csv, database queries, etc) with varying formats and potentially missing or mislabeled information. Experience and familiarity with various data types, including images, tabular, unstructured, and text. Experience working with other subject matter experts to actively solicit information needed to harmonise datasets into machine learnable form. Experience in document mining and processing diverse data sources. Experience working with several large public scientific (preferably biology-related) data sources is a plus. Experience with validating the data accuracy and quality of data by cleaning, shaping, analyzing, normalizing, and conforming it to existing models and vocabularies. Strong interest in the latest relevant literature and application of ML and data science to the biological sciences, is a plus. Understanding and working vocabulary about common machine-learning concepts (training sets vs. test sets, over/under fitting, bias, annotations, feature extraction, RAG, LLMs, classifiers, and so on) Excellent English-language oral and written communication skills. Proactive communication habits: asking questions and seeking clarifications when necessary. Desirable Requirements: BS in Computer Science, Informatics or similar, or equivalent practical experience. Fluency in English
