Data Scientist I (Global Onco)

Job Description

Understand sophisticated and critical business problems from Country/Regional/Global business functions, formulate coordinated analytical approach to mine data sources, employ statistical methods, machine learning & deep learning algorithms to discover actionable insights and automate process for reducing effort and time for repeated use. High agility to be able to work across various business domains and divisions (Oncology, GenMeds, Sandoz). Able to use business presentations, impactful visualization of ML results and contextual storytelling to translate findings back to business users with a clear impact. No direct team management. Work on a variety of business applications including but not limited to: Customer Segmentation & Targeting, Event Prediction, Propensity Modelling, Churn Modelling, Customer Lifetime Value Estimation, Forecasting, Recommender Systems, Marketing Mix Optimization, Price Optimization Be involved in technology research, capability building across newer technologies and tools in Machine Learning ecosystem at scale. Design, run and analyze A/B and multivariate hypothesis tests sought at optimizing customer and patient experience. Also, pick up new skills and technologies necessary on the job Articulate solutions/recommendations to business users. Works with senior data science team member to present analytical content concisely and effectively. Develop automation for repeatedly refreshing analysis and generating insights Collaborate with globally dispersed internal stakeholders and cross-functional teams to solve critical business problems and deliver successfully on high visibility strategic initiatives. Understand life science data sources including sales, contracting, promotions, social media, patient claims and Real-World Evidence Project manage own tasks and works with allied team members; plans proactively, anticipates and actively manages change, sets stakeholder expectations as required. Independently identifies research articles and reproduce/apply methodology to Novartis business problems Ability to work independently, demonstrate initiative and flexibility through effective and innovative leadership. Attention to detail and quality focused, excellent interpersonal and communication skills, innovative, and collaborative behaviors & strong can-do orientation Assist in hiring of new associates for data science roles. Comply to all Novartis operating procedures as per legal/IT/HR requirementsMinimum requirements What You ll bring to the role: Education: PhD or Masters (or Bachelors from a top Tier University) in a quantitative discipline (e.g. Statistics, Economics, Mathematics, Computer Science, Bioinformatics, Ops Research, etc.) 7+ years of proven experience in Data Science. In case of PhD, 5+ years post qualification experience. Experience in commercial pharma would be an added bonus. Extensive experience required in: Statistical and Machine Learning techniques including but not limited to Regression (esp., GLM, non-linear, etc.), Classification (CART, RF, SVM, GBM, etc.) Clustering, Design of Experiments, Exploratory Data Analysis, Statistical Inference, Feature Engineering, Time Series Forecasting, Text Mining and Natural Language Processing (NLP). Crafting and deploying ML modeling and prediction pipelines Good to have skills: Stochastic models, Bayesian Models, Markov Chains, Dynamic Programming and Optimization techniques, Deep Learning techniques on structured and unstructured data, Recommender Systems (content and collaborative filtering), etc. Tools and Packages: Good command over Python, R, SAS. Strong coding skills with the ability to write high-performance code in Python; exposure to PySpark, Tensorflow. Proficient with SQL and Hive. Exposure to DataIku and UI interface tools like R-Shiny, Streamlit, etc. desirable. Exposure to AWS and ML Pipelines on cloud desirable,

Keyskills :
machine learningpythondata analysissqlanalyticsdesign of experimentsexploratory data analysisnatural language processingtext miningtime series