Building sustainable & long-term analytical solutions.

Data Scientist position in the Merchant Services area, Customer Intelligence and Campaigns.Key activites:- Retention: Building proactive retention ML model and its productionization.- Data engineering: Support of sustainable data flows and their execution, support in design and architecture of data flows on Hadoop- Data extracts: Ad-hoc data extracts/data analyses to provide answer(s) to the business unit- Setting up a generic data science environment, which allows data scientists to build models and ML products in a productionized fashionTools:- Python, R, Spark, Hive

Data scientist position in the Global Analytics department with main focus on big data platforms and their usage in Data Science area for business purposes.Key projects: - Fraud: Acquisition and cleaning of the source data for ML purposes, data preparation, data wrangling, building ML models, productionization of ML models for real-time scoring, optimization of ML models and data science processes, automization of DS processes (offline champion-challenger setup), investigation, visualization and skeptical ways of evaluation of both models and results, continuous communication and alignment with the stakeholders and business units-AMLTool package:- Big data: HDFS, Hadoop, Spark, Hive (Beeline)- Programming: Scala, Python- DS: Python, JupyterResearch areas:- Bayesian optimization

A collaboration between University of Copenhagen and Danske Bank with focus on credit card fraud detection, based on standard and more advanced machine learning algorithms. The output should be a feasible combination of models, which in combination accurately predict the probability of fraud transaction-wise, including the explanation of the score and the models should be applicable to production if needed.

Student data scientist positionKey activities: - big data work: including data analysis, data moving, data cleaning, sanity checking, data quality assurance, data transformations - feature engineering; spark, scala, python, R - modelling: imbalanced problem modelling, classification and regression models, model evaluation, model performance; spark, scala, python, R, - production: real-time analytics, productionizing models; python, scala - R&D: innovative modelling techniques, model evaluation research, Spark and Python development - communication and cooperation; LaTeX, TeXMain projects: - tools investigation - fraud detection - deep learningTools and platform: - Apache Hadoop : hive, spark (scala), ambari, yarn - SQL - Python - Scala, Spark, R - statistical software: R, Wolfram Mathematica, Matlab - version control: git - documentation: LaTeX, TeX

Acquisition of new clientele, consulting, complex analysis of needs and matching expectations, customer service, retail