Machine learning in a young start up does not look like a Kaggle competition. Data science projects start with a more extensive roadmap than dataset. In the absence of data, subject-matter knowledge makes heuristic solutions a tempting first step for all stakeholders involved. While rules-based algorithms are not the glamorous side of data science, they need not be a dead end and can form the basis for increasingly sophisticated labeled data. In this talk, we propose a path to iteratively bootstrap a supervised machine learning model out of heuristics and demonstrate its potential on the basis of N26 projects.
N26 is one of Europe’s first and fastest growing mobile banks, offering the same services as the traditional bank, but enabling the customer to manage their account directly from their phones. Their vision is to offer customers a way of banking that is easy, personal, mobile and realtime. This means, for example, to get push notifications for any transaction that one makes, to be able to set and change one’s daily payment and withdrawal limits, to immediately lock one’s card if it gets lost, reset one’s PIN and many other useful features.
When the mobile bank launched in private beta version in October 2014 at TechCrunch Disrupt London, the bank N26 was still a fintech startup, called Number26. Since then, N26 has attracted over half a million customers, made their services available in 17 European countries and plans to launch in the US by mid 2018.
The event took place on December 1st, 2017.
Find Jeremiah’s slides on SlideShare.