Data Science Summer Internship

MSIG USA•Warren Township, NJ

20h•Hybrid

About The Position

MSIG USA is seeking a Data Science Intern to support the design, development, and evaluation of predictive models that inform business decisions across insurance / financial services use cases (e.g., risk, pricing, propensity, claims). The intern will work alongside experienced data scientists and contribute to real analyses using production-like, real-world datasets. The internship aims to build strong foundations in applied machine learning, experimentation, and stakeholder communication.

Requirements

Python (Pandas, NumPy; plus scikit-learn or similar ML libraries).
Solid understanding of statistics and probability (distributions, confidence intervals, hypothesis testing).
Working knowledge of machine learning concepts (classification, regression, model evaluation).
SQL: ability to query, join, and aggregate data from relational sources.
Currently pursuing a Bachelor’s or Master’s degree in Data Science, Computer Science, Statistics, Mathematics, or a related quantitative field.
Coursework or project experience in machine learning/statistics and data analysis.

Nice To Haves

Experience with Jupyter / VS Code and version control (Git) is a plus.
Strong analytical/problem-solving skills and curiosity about “why the data looks this way.”
Ability to explain technical ideas in clear, simple language to non-technical audiences.
Attention to detail and comfort working with messy, real-world data.
Reliability, ownership, and willingness to ask questions early.
Experience with Git-based workflows (pull requests, code reviews).
Familiarity with model monitoring concepts or production analytics practices.
Exposure to insurance, claims, underwriting, pricing, or financial datasets (helpful but not required).

Responsibilities

Support predictive model development for use cases such as risk, pricing, propensity, and claims.
Perform data exploration and feature engineering using Python and SQL.
Clean and prepare datasets from multiple sources (e.g., EDP, PRS, finance/claims systems).
Run experiments and model evaluations, including train/validation splits, performance metrics, and back-testing.
Translate results into clear summaries and visuals for business stakeholders.
Contribute to documentation of datasets, models, and experiments within internal repositories.