Data Scientist

In this role, you will:

• Address model drift and performance of business requirements as part of recurring operations, drive quantitative model maintenance prioritization, and inform stakeholders.
• Understand broad spectrum of Machine Learning and statistical models, impactful features from structured/unstructured data.
• Understand and have experience in models that help automation, getting insights, make smart decisions, are able to meet the desired KPIs post-production and predictive modelling in Python, Azure (optional), Insurance(optional)
• Stay abreast of the developments in the rapidly changing field of AI and machine learning
• Gain deeper understanding of Chubb's business and develop expertise on multiple lines of business
• Collaborate with business partners and peers within the organization to understand the machine learning models and develop robust model monitoring solutions that ensure data quality and analytic accuracy.
• Execute all aspects of analytics model monitoring including exploratory data analysis, data processing, and model monitoring.
• Ability to identify trends and outliers in model results and analyze drift and impact to business value.
• Research, recommend, and implement statistical and other mathematical methodologies appropriate for model monitoring.
• Create/Maintain excellent working relationships with business partners across the Chubb organization
• Provide supporting documentation for the models being monitored including documentation of methodologies used and data issues encountered.

Qualifications

• Required:
  • Some experience utilizing model monitoring methodologies.
• Understanding of data mining predictive modeling, and ML concepts/ Probabilistic Models, Ensemble Techniques, Hyperparameter Optimization, machine learning models/sound knowledge of Regression/ Logistic Regression, Random forest, XGBoost, SVM, Clustering, statistical modelling (GLMs in particular), etc.
• Proficient in Python with strong capabilities in data analysis/modelling
• 4-8 years programming experience in using Python libraries / machine learning libraries.
• Strong Programming experience in Python.
• Strong Programming experience in SQL.
• The ability to multi-task, learn new things quickly, and demonstrate excellent problem-solving skills.
• Preferred:
  • Working knowledge/familiarity with Git version control.
  • Practical experience with MS Azure and Databricks.
  • Experience in architecting and consuming APIs at scale.
  • Desirable to have experience in visualization tools like Qlik, Power BI, Tableau etc.
  • Experience with Text Analytics and Natural Language Processing. Experience with BERT models/understanding of vector embeddings and operations on unstructured data.
• Nice to have:
  • Advanced knowledge of model tuning, evaluation, and operationalization.
  • Experience with Deep Learning libraries (Tensorflow / Keras, PyTorch, MXNet, etc.). Experience with Azure ML/ MLFlow.
  • Experience with at least one other programming language (Python, R, Julia, Scala, Go, Java, C++).

Education:

Bachelors degree in Statistics, Economics, Data Science, or related field. Advanced degree a plus.