Data Scientist | Sunnyvale, CA, USA

Data Scientist Calling all innovators - find your future at Fiserv. We're Fiserv, a global leader in Fintech and payments, and we move money and information in a way that moves the world. We connect financial institutions, corporations, merchants and consumers to one another millions of times a day - quickly, reliably, and securely. Any time you swipe your credit card, pay through a mobile app, or withdraw money from the bank, we're involved. If you want to make an impact on a global scale, come make a difference at Fiserv. Job Title Data Scientist What does a great Data Scientist do? The Risk Analysts in EPayments are the data scientists responsible for the identification of fraud behavior patterns, the estimation of predictive models and the installation of policies to reduce losses for our client banks. Our analysts are responsible for the full predictive modeling cycle, including data preparation, model estimation presentation to clients, installation of policies and performance tracking. This position is an excellent opportunity to learn and develop modeling skills for the e-payment sectors, work in a large-date environment and apply statistical forecasting techniques. It also provides a high degree of interaction with internal decision makers and external stakeholders in the industry. As a Risk Data Scientist, you can look forward to: • Using statistical techniques to examine payment data provide measures for outlier identification on key measures such as transfer velocity, ip-distance profiling and device recognition. • Creating the predictive measures for the patterns and collecting the information using SQL on ORACLE and Python/Spark data structures. • Estimating predictive classification models to separate legitimate from fraudulent behavior, using XGBoost, LighGBM and neural network libraries. • Installing the predictive model code in our systems and collecting performance both in ORACLE and Python and creating the policy for actions to be taken (manual review, live challenge using SMS, suspension etc...) • Interacting with heads of the fraud-strategy departments at the client banks. Our analysts present the policies, document the models and negotiate approvals and timelines for implementation with the stakeholder at the bank.

: • Thorough understanding of statistics, confidence intervals and some background on CHAID, statistical distributions and estimation methods such as regression, binning confidence interval etc. • Must have hands-on experience collecting data in SQL from data systems and summarize the attributes for the data in counters, sums, histograms and frequencies. Summarize information for distributions and finding quantiles and running totals. • Hands on experience processing data in Python: Should be able to read a csv file, import the data, run counters and summary distribution, aggregate totals and do exploratory analysis on a data set. • Should be able to estimate classification/ predictive models using the open-source library in Scikit learn for Python. Measure the goodness of fit and compare them across alternative models, such as ROC, gains charts. • Should be able to measure changes in underlying distributions, and compare between different populations or policies. • Document the analysis and present it verbally providing insights on the factors that are most relevant for the model and how can it be used to predict/manage a policy.