American Express - Consultant - Decision Sciences - RENEW - IIT/NIT/BITS/IIM

American Express - Consultant - Decision Sciences - ReNEW

#ReNEW

A Second Career Initiative for Women by American Express

American Express encourages you to come back in the game!

Whether it is your employability at American express at the end of your project tenure or your testing grounds for sharpening key skills, there is plenty in store for you to experience and make most of the 6-month journey.

Eligibility: (- who are on a minimum of 6 months of career break and in addition hold at least 4 years of relevant work experience)

Business Overview:

Global Customer Management Decision Science team is responsible for ownership of Customer Management Models, namely, Card member Decision Support System (CDSS) and Total Structural Risk (TSR). The portfolios under the purview of the team are US and International Consumer and Small Businesses and Corporate Clients/Card Members. We use their outputs to impose downstream controls that prevent extensions of credit beyond the Cardmember's ability to repay. Examples of downstream controls are declining high-risk point-of-sale transactions, establishing or modifying limits on card exposure and, in extreme scenarios- where the anticipated risk is unacceptably high- proactive cancellation.

Over last 20 years, the customer management models have grown increasingly accurate and powerful through continuous innovations in analytical techniques and data. Examples are the use of machine learning and the creation of composite variables that extract a maximum of informational power from standalone variables. In 2016, the team rolled out GBM based customer management models for both international consumers and small businesses as well as corporate clients and card members across the globe which drove $36MM in PTI

Project Name:

Build and manage the end-to-end launch of the Next Generation Probability of Collections Model

Project Details:

Build the next generation collections model that significantly enhances the existing model on multiple dimensions:

Potential 2 stage construct (Probability of Contact and Probability of Payment conditional on contact)

Enhanced use of:

Rich transactional data ( including type of spend e.g. spend at lawyers) and external data (e.g. tax liens from Enigma)

Customer contact history and interaction with Amex (e.g. call responses, MYCA logins)

Income and asset information (change in income from tax, home value/equity/sale)

Segmented dependent variable definition based on charge versus lending.

Pre-requisites:

Education: Engineering/Science/Economics/Statistics /Maths/MBA(Finance) background .from premium institutes (like IIT- s, BITS, NIT's & IIMs).

Overall 4+ years of experience including 2-3 years of experience in Data Science.

Modeling experience in the financial domain in preferred.

Hands on Programming Experience is required.

Experience in Machine Learning techniques (preferred).