Who are the faces behind the software for Internet competitor monitoring? In our blog, we will be introducing our colleagues. blackbee co-founder and CIO Sabine Maßmann and data analyst Sandra Engelhardt made the start, and now our co-founder and CTO (Chief Technology Officer) Dr. Hanna Köpcke has her word. The big data expert tells us how she is further developing blackbee, for which she is currently writing a research proposal, and why a measure of curiosity is part of the job.
Which tasks do you perform?
Currently, I’m busy preparing for the new topics that are being addressed in the field of big data matching. Additionally, I help my colleagues so that the matching processes work (customer data is compared with that of other providers on the Internet and the software combines various product attributes for this: Author’s note).
Besides technical management, you are also responsible for research and development. What is being researched and developed at blackbee?
We want to start a new project with the DFKI (German Research Center for Artificial Intelligence) and I am currently writing the application so that we get the necessary support. It’s about so-called deep learning – or deep neural networks that we want to use to enable better big data matching algorithms. First of all, we will be developing this together with the DFKI and after that the goal is to integrate this procedure into our processes in blackbee.
What would be gained from this deep learning matching?
So far, we are still working with the matching method that I developed back then. Deep learning is more akin to the abilities of a human brain. It consists of several layers, of networks, and selectively links various sets of information with one another. This has only been possible for a short time, since the processing power of computers has now improved worldwide.
What did your team and you work on lately?
We are working on solutions that allow us to match images in big data matching. This is a major topic, especially in the fashion sector. There are a lot of black T-shirts from every manufacturer, which are called simply “T-Shirt, black”. It is difficult to differentiate these through product attributes. The customers look at the pictures and read less through the descriptions of these clothes.
In this area, we have just made our first attempts with an existing customer. In general, the recognition of objects in images is currently a major topic in artificial intelligence research.
Dr. Hanna Köpcke developed the big data matching process, with which blackbee works today, and is also constantly improving it. She is currently planning a project together with the German Research Center for Artificial Intelligence.
What do you value most among your colleagues? What do you look for in applicants?
Whoever applies for a matching position first receives a matching task and then I look at that candidate’s approach. What are the ideas? What are the proposed solutions? No previous experience in machine learning or similar is needed, but applicants should be inventive and want to get to the bottom of the problems. I then take a look at their approach. How is the code programmed? Is it clearly arranged? Do I understand it at all? Has the applicant written comments in the source code so that it is easier to understand? Have they developed test cases for their code?
If that works out, the applicant is then invited to interview.
When did you realise that the world needs blackbee?
I did not do it alone. I came to Leipzig to the Chair of Mathematics and Computer Science to gain my doctorate in Artificial Intelligence. In that time, we three – Carina, Sabine and I – developed blackbee and quickly realised that blackbee would be well received in the eCommerce market.
Were you already able to gain your first customers then?
We were quickly in contact with interested parties. At one of our first trade fairs, we got to know companies that still count among our customers today.
What advice do you give women in the tech scene to be successful?
Keep on fighting. Be brave and remain curious! At the beginning of my studies I thought to myself, Boy, they can do everything here! And often it turned out that only hot air was behind it all. In my experience, women in IT are rather shy and only speak out when they are 100 percent sure. There is no real reason for this.
Knowledge and programming are not always the most important in IT. It also involves curiosity.
Do you need further information on the topics of competitor analysis, market analysis and price management on the Internet? Contact us now – we look forward to your message!