Interview by Thomas Peham
March 10, 2015
Photos by PAYMILL
Developing innovative products in fintech
In his role as VP of Engineering at the European startup PAYMILL he's working on some major innovations in the finance industry.
Before joining PAYMILL Christian Braun worked as freelancer a couple of years and later founded his own company.
Share this interview:
Christian Braun, thanks for your time! Can you describe yourself in a few sentences?
I started my own company in 1996. At that time my dream came true to have an own company. I sold it later on in 2001 to another company and afterwards I started to learn everything I’ve build before in practice. I then started to study computer science at university. Afterwards, I wanted to work in computer science and started at a print company where I was responsible for the software development and project management team. It was always my dream to develop stuff and create useful software.
Now, at PAYMILL, my job is about developing a great product in the fintech area. It’s more about managing people now and less coding on the products itself. I can create the IT vision of PAYMILL and I can rely on a strong team executing these ideas.
You mentioned that you sold your own company in 2001. What was the company about?
It was the e-commerce company called, "NET Order GmbH". It was an online shop which had a direct connection to a ERP system called Olymp. There was no other online shop at this time who stored orders in a ERP system. Everything was generated from the ERP system which was quite a huge deal at that time. This is why we sold quite a lot of these e-commerce shops.
How would you describe a typical day in the life of Christian Braun today?
I’d say that there are a lot of meetings and phone interviews for recruiting new developers. A lot of time is spent on managing priorities for the development team and having vision meetings for the next months.
Has there been any moment which you would call a big influence for your career you choose later on?
My grandpa had a company which was sold a long time before I was five years old. He died very early but in that matter it was always about the “company”. I think that was something which I wanted to achieve too.
And of course, I was interested in all the new stuff in my childhood. My first computer was a Commodore c 64. I got that while my mother was on holiday and I convinced my dad to buy me one. That was the start of developing stuff.
Was there any "Aha" moment where you knew that technology is really your thing?
I think it was likely this first computer. It wasn’t only about playing games, but also having a look on how can you hack that software. Exploring the basics of a computer was really the interesting stuff back then.
Can you name one thing you really like about working as a VP of Engineering at PAYMILL?
I like to work with people and it’s also great to have a good team like I have. When everybody has different opinions and is really open-minded not to follow the mainstream every time. Also, the combination of the teams we have. Not every developers has a university degree. They are also coming from other industries. It’s really important to have such an interdisciplinary team.
PAYMILL’s developer friendly REST API lets you accept credit and debit card payments directly in your website and mobile application. The API gives you full control of your checkout so you can deliver a user experience optimised for higher conversions. And with heavy-duty security measures already built-in – you can rest easy knowing your data secured. Further information at paymill.com
Do you see any challenges in your team's structure or in your engineering department which you face at the moment?
Of course, you have different challenges as an engineering department. One major challenge is the question of how to hire good developers, because everybody wants them. You also have a challenge that you have enough developers in order to work on a lot of product improvements.
Then there’s the challenge that you have to work simultaneously on different products and ship them and offer them to customers. Further on, the system architecture is a big challenge too.
For example, working on a smooth API, so that people are going to use your products. We got a lot of great feedback from our customer support that people like our API for example. Of course, another challenge is being one step ahead of everybody else in the industry. You don’t want to miss some specific points but you’ll have to think differently in order to compete on the market.
Is there any tool or any software you use on a daily basis and you'd like to recommend to others?
Yes, there are different software products which I use on daily basis. There is JIRA and confluence from Atlassian which we use in our bug tracking and product management process. We are also managing our process with some other add-ons for JIRA. Our scrum sprints take place with greenhopper for JIRA. We are also using that tool for change requests. Every ticket which is managed in JIRA is handled as a change request.
We estimate change requests with t-shirt sizes. Then the management can get a clear overview about costs and priorities.
We are putting priorities on that, then estimating it with t-shirt sizes so that the management can have a feeling about estimations. It starts from XS and going up to XXL, so that the management can have a feeling how much it will cost and how much it will impact the product.
We use Gerrit for that which is a really cool tool. We’ve also included it into the JIRA workflow. Before any code is sent to QA for testing, a code review is done where every developer can give a plus 1. So different developers are giving feedback with plus 1 or minus 1 and then the code is going back to the developer in charge. He then needs to do the stuff on which he got feedback from the other developers.
What we also have is Jenkins for continuous integration. In that part, the operations team also uses Puppet, which is really a good tool to deploy stuff on distributed servers.
We also use Nitrate for managing our test cases stories. That's a really interesting tool. It's nothing automated - we use it to write the test cases there. We use Cucumber (Ruby) for system tests (automated tests), and we are switching from Cucumber now to Behat. Behat is the PHP framework and very useful for functional or system tests.
Is there any advice you would give yourself as a 14-year old?
I’d only say: take stocks from technical companies like Apple, Microsoft or Google. I think I would have done everything else in the same way. In the century I grew up it was more fun to play with new things and tools. You had the ability to do so much stuff, because the power of the processors wasn’t as good as nowadays. The graphic interface was not as good.
If I’m a 14 year old now, I think I would also give myself the advice to play with non-digital stuff too. I think it was really helpful, not to have a mobile phone with me when I grew up. That was a really nice thing, because nobody else cared when you were not reachable at a time.
Is there a project or technology you'd like to explore?
I’d like to dive deeper into mobile. My diploma thesis was about Android 0.5. We have a mobile SDK at the moment, and working on mobile would be really interesting because it will be a game changer in the next decades. Everything looks quite nice nowadays and Android Pay (Framework) for example got announced, Samsung Pay was also announced now as competitor to Apple Pay. I think there is quite an interesting market there, even with iBeacon from Apple. I think what really is interesting to me is the influence of these things and how PAYMILL influences the “good old industry” of banking. To ditch old requirements and start some new innovations in finance. That's also one of the reasons I work at PAYMILL.
You mentioned that you also worked as a freelancer for a couple of years. Is there any important lesson you've learned during that time you'd like to share with us?
I think one important lesson at freelancing I’ve learned at that time is that you don’t have to work 24/7. That's not good for your health. I think you should spend some time on new things which free your mind. Take your time to explore new things and try to do things differently and convince the customer that it is also helpful for him.
What are the biggest pains in software testing?
Yes, of course there are some challenges. If you are writing system tests, you should be keeping in mind that the tests should be scalable. That's really important. If you have continuous integration, then you need to wait until the next one deploys them.
The QA members need to be part of the team right at the kickoff of the product development. They need to be involved directly at the beginning in order to develop better tests.
Additionally it is essential that you need to understand everything if you write test cases before you start to code tests. Having an understanding of the stuff you are just building is really important - more or less for everything is that important ;). Can I conduct the test? Can I involve my QA team? Can I conduct a high quality test? PHP unit tests are really a must have. Functional or system tests are a must-have too. For us - at PAYMILL - unit testing must be one hundred percent perfect. System tests can be conducted a bit later on, but are also a must-have before deployment. We give developers the time to conduct well-prepared tests.
Sometimes, it can be hard to convince the management team not to say, "We skip the test", in order to have a product ready earlier but winnning time by skipping tests is a misbelief. But as a payment processing company, the most important thing is when all transactions are processed successfully. That's really important for us. Thus, conducting must-have tests is a key to product success.
What would you describe as your main passion about building software?
At the beginning of my career, I built products for a certain life cycle, like an architect who is really happy when he sees the same building after 50 years. That was my thought. Later on, I discovered that this isn't true, because all the developed stuff which I developed in the 90ies has changed but the ideas still exists. For example a ticketing system I built at Bytesteps AG was then bought by the Telekom. It was a ticketing system which is even now, I think, used by FC Bayern Munich. That was designed by me at the beginning, but it changed a lot of course. I think another motivation is that you design and build stuff that other people enjoy using for a specific time. Make something which is really easy-to-use and has a clear meaning. For example that was the case when we built the API for PAYMILL. We looked at it and said that we wanted to have something that is really straightforward and really easy-to-use. You don't have to read 150 pages in order to implement that stuff right away. I think that's really important. That's the key motivation for me actually.
What do you think will be the major challenges in software engineering in the next three years?
There will be a couple of challenges. The one challenge is to have an environment which you can develop on mobile and on desktop. I think that's a challenge to get mobile and desktop together again, and to develop a all-in-one solution. Right now it's not manageable to develop this all-in-one solution. At the moment you develop for every platform. I think that's really a challenge.
Then, cloud computing is a huge topic. Of course, there are some regions in different countries that have no internet connection. If everything will be developed in the cloud we do face a major challenge. And if you think of cars driven by computers and all that connected systems.
One last question, do you have a couple of resources, books, blogs, or videos you'd like to share with us concerning web development?
I read a lot of articles on heise.de. There are some blogs I use concerning AngularJS, which is a really cool framework. I recommend reading posts from Pascal Precht. I also have some payment blogs which are really great like letstalkpayments.com or some Linkedin groups. And I also read a lot from techcrunch.com or gizmodo.de.