Meet the MANTAs: Andrej Jurčo, Software Engineer and Python Team Lead
In this Q&A, software engineer and Python team lead Andrej Jurco shares his thoughts on what it’s like to swim with the MANTAs.
What inspired you to join MANTA?
I first heard about MANTA a few years ago, when Lukáš Hermann, MANTA’s VP of Technology, gave a presentation about what the company does, what issues they are dealing with, and why it results in a unique offering. The idea of taking metadata from some technology, applying a fair bit of magic, and getting a data lineage graph seemed very cool.
Additionally, thanks to MANTA’s cooperation with major universities in Prague, it was possible to work on my thesis at MANTA, and I thought that it would be interesting to learn a bit of that magic. Another appealing aspect of MANTA was that when I joined, the company was still relatively small and everyone I met was very nice — and they still are, even though we have grown so much since then.
What is your team currently working on, and what projects get them excited?
We are currently working on a scanner that uses static analysis to create data lineage from Python scripts. It is a new undertaking and, together with our scanners for Java and C#, a unique thing. Because of this, we cooperate with universities to conduct research with us in this field. It is a very exciting job — you never know what problem you are going to come across.
Because the scanner is so specific, it is a very creative job and we must do almost everything from scratch. There is not usually another solution that we can simply use. The research nature of our job is probably also what gets the team excited. If we realize that some approach that we’ve been using for a couple of months doesn’t work well enough, or if we come across a language feature that is very difficult to process, the team gets excited that a new solution must be created. Simply said — the harder the job, the more excited the team becomes.
What advice would you give to an engineer who wants to join your team?
The best advice is probably that they should not be afraid to try and fail. Everything we do in our team is completely new and there is almost always something that does not work and creates a chain of follow-up issues. Often, you find out that you started in a completely wrong way and must try to solve the problem with a different approach. This is normal, and without the risk of making a small mistake, there is rarely any progress. If people dwell for a long time over everything that does not go their way, they are not going to have an easy time here.
What’s next for you and your team?
There is always something to work on in the Python team and I don’t think that we are going to run out of features to deliver any time soon. The language is developing rapidly and there are always new features that need to be delivered, new libraries that need to be supported, and new Python-using technologies that are going to need new scanners.
Currently, we are working on supporting new libraries, such as PySpark or Pandas, as well as new technologies that utilize Python, such as AWS Glue or Databricks. Additionally, we are experimenting with new approaches for faster and more precise results.
