Data Scientist - Python/R - Credit Risk Domain - BFS

Data Scientist - Credit Risk - Python/R

Key Responsibilities:

- Partnering with clients and internal business owners (product, marketing, edit, etc.) to understand needs and develop models and products for organisation(fintech) business line.

- Good understanding of the underlying business and workings of cross functional teams for successful execution

- Design and develop analyses based on business requirement needs and challenges.

- Leveraging statistical analysis on consumer research and data mining projects, including segmentation, clustering, factor analysis, multivariate regression, predictive modeling, hyperparameter tuning, ensembling etc.

- Providing statistical analysis on custom research projects and consult on A/B testing and other statistical analysis as needed. Other reports and custom analysis as required.

- Identify and use appropriate investigative and analytical technologies to interpret and verify results.

- Apply and learn a wide variety of tools and languages to achieve results

- Use best practices to develop statistical and/ or machine learning techniques to build models that address business needs.

- Collaborate with the team to improve the effectiveness of business decisions using data and machine learning/predictive modeling.

- Innovate on projects by using new modeling techniques or tools.

- Utilize effective project planning techniques to break down complex projects into tasks and ensure deadlines are kept.

- Communicate findings to team and leadership to ensure models are well understood and incorporated into business processes.

Skills:

- 3+ year experience in advanced analytics, model building, statistical modeling, optimization, and machine learning algorithms.

- Machine Learning Algorithms: Crystal clear understanding, coding, implementation, error analysis, model tuning knowledge on Linear Regression, Logistic Regression, SVM, shallow Neural Networks, clustering, Decision Trees, Random forest, Boosting trees, Recommender Systems, ARIMA and Anomaly Detection. Feature selection, hyper parameters tuning, model selection and error analysis, ensemble methods.

- Strong with programming languages like Python/R and data processing using SQL or equivalent and ability to experiment with newer open source tools

- Experience in normalizing data to ensure it is homogeneous and consistently formatted to enable sorting, query and analysis.

- Experience designing, developing, implementing and maintaining a database and programs to manage data analysis efforts.

- Experience with big data and cloud computing viz. Spark, Hadoop (MapReduce, PIG, HIVE)

- Experience in risk and credit scoring domains preferred