Travelpayouts Blog

Meet our team: Artem Makhortov, system administrator

Reading time:  6  min.

Artem Makhortov is one of the “Horsemen of the Apocalypse” of our system administrators team. He told us why his colleagues view technology in a different way than developers, why it is more reasonable to rent servers instead of buying one, what system administrators do with unscrupulous users, how they sleep hugging a laptop on duty and why aliens should drop by Phuket.

Before Travelpayouts

  • As a child, I wanted to be a discoverer. Back then, I believed that I could bring something useful to this world.
  • My friends and I worked as supervisors for various street promotional campaigns in the early years of my study in the university. Once, we presented a car window cleaner in a huge supermarket parking lot. The guys disturbed the passers-by, and I had to take them by car from one place to another. Approximately on the fourth day, I gathered all the taxi drivers together and we sat all day and smoked a hookah right in this parking lot. It was funny, although we were not paid much for this work.
  • I am a financial broker, I graduated from the Financial Academy under the Government of the Russian Federation. Six months of working in this sphere were enough to understand that the job was not my cup of tea. During my studies at university, I started working in technical support, it was a large corporation, then a bank and finally in administration.

My role at Travelpayouts

  • There are three system administrators in our team, and we have approximately the same range of responsibility. Our head administrator Evgeniy Kozlov, who we call uncle Zhenia, is responsible for sorting out the worst disasters that could cause a loss in profit.
  • Our team supports the whole infrastructure: Aviasales, Travelpayouts and all the additional services such as analytics, marketing and different internal services for developers.
  • Our vast monitoring system Zabbix takes metrics from the whole infrastructure. There is a function of sending an alert by mail to us if something goes wrong. In this regard, we assign shifts accordingly: each of us works for two weeks and then has one month off. While working, we always turn on all the alerts on the mobile phone and wait for a signal, even when sleeping. We even take a laptop everywhere we go: for lunch, to the restroom, cinema or a party. When you receive a failure alert, it’s time to fix the problem. I’m lucky, I never had any serious problems during my shifts.

Work process

  • If our sites open and work as you expect, it’s a job well done.
  • Usually, we use our main software Zabbix for tracking and monitoring. We changed the infrastructure and converted to a new virtualization system, so we started to introduce Prometheus. We keep up to date, so to say.


  • We have a different view on technology in comparison to developers. First of all, we think about how to use innovation in the future, how it will be developed later, how to implement it and make it work properly. Software engineers will always use the newest, the coolest and the most advanced. But, it isn’t always good. Our company is rather advanced and we try to use the latest technologies such as kubernetes and docker, and new languages ​​such as Golang. All these innovations are stylish, fashionable and pretty new. Little by little, we switch to these things, and in theory, it will take some weight off and help with the development.
    We do not buy servers because we rent them from one hosting, which I will keep secret, as to not advertise it. In fact, renting servers gives you more flexibility to make decisions regarding infrastructure, while purchased equipment needs maintenance. We are located in Phuket, which is quite far from the points of our presence, and hiring an employee for each one of these points is not reasonable. Also, we can easily scale which helps us follow the latest trends as the requirements for the “hardware” component change constantly. For example, we recently were looking for special servers with an increased quantity of CPU cores\threads for new containers and negotiated it with our hoster just because everything works better on this “hardware.”
  • Luxembourg, Amsterdam and Moscow are our three points of presence. The main ones are in the first two cities. If something goes wrong in one data center (for example, our server burns down, there is a lack of electricity or the internet connection is down for a long time), there is always a second data center with the same services and the same configuration. We can easily switch and continue working. Some special marketing projects, of which the stability and security are not so important, are located in Moscow. It is cheaper and the users are closer.
  • When I joined the company, I saw their “moving” process from one host to another. Sometimes, it just so happens that you need to move a server from one place to another inside one hoster. We try to make it more or less non-obvious to our users and partners. But from time to time it is impossible to hide this process, and this is when you see windows like “statistics are temporarily unavailable” on the site. Usually, it means that an internal infrastructure improvement is going on, I hope that users will understand.
  • It goes without saying that the growth of the company has had a significant influence on our work. We remade the database and divided it into different parts, improved the productivity of our servers and our programmers rewrote the old programs to make our work better. We try to do our best and we are good at it.

Work issues

  • Once, there was a situation when one of our hosters turned off a lot of servers, and Aviasales and Travelpayouts lost the main functionality. It was impossible to open the control panel on the Travelpayouts website and enter a search request on the Aviasales website. Everybody suffers from it. So, when such difficult situations arise and negatively affect the business, it makes our heads spin.
  • We’ve never had any serious DDOS attacks. But sometimes there is that “smart” person who tries to bend the rules, whether it is an unscrupulous partner or users who check our database for subscriptions. We ban them straight away. It doesn’t influence the core functionality, but it affects the quality of work. Such incidents might make our work take longer because the requests take longer to process as well.

Work environment


  • I would evaluate the technical proficiency of the Travelpayouts team at 7 or even 8 out of 10 points.
  • I already knew some people from Aviasales, which is why it was easy to fit in. I was afraid of Phuket more than new people, and I can’t say I’m a fan of tropical countries. Before that, I had never been to Asia, except China and Singapore, though it is another matter. But working in Asia is “hardcore” for a city person. So it was quite difficult for me to get used to the local rhythm.
  • I worked in a huge corporation with 3,000 people, and then in a small company with 10 people, so I have something to compare my environment with. I like the way our communication and sharing expertise work.  If it was a big company, it would be difficult to achieve such a level of communication.
  • We have our internal messenger Slack, where everyone starts asking a newcomer a lot of ridiculous questions, not related to the work, like “what spice are you?” I think it is cool to have a team language within the company to communicate about more than work issues. This cultural code is always interesting, and it exists in a few companies.

About achievements

  • Our team constantly improves the office life. We have a tradition of gathering for demo presentations to show the interesting things each team has done during the week. Recently, we made an online demo broadcast. It means that even if our colleague works remotely, for example, Andrey Safronov, he can still be a part of our discussion.

About motivation

  • My motivation is to go to a bar with my friends. Self-motivation and other similar things are not for me. I find it weird when you need the motivation to do something that you like.

Things I’d like to improve

  • In IT, you always learn something new. You have to figure it all out whether you want it or not. If it is necessary to master new technology, no problem, we will do it. Therefore, “training yourself” is a permanent process.

Let’s get personal

  • I can not describe my job as fabulous or fantastic, but it is an interesting thing that I am really keen on doing.
    If aliens arrive and ask where to go on vacation in order to understand the human race better, I would advise them to visit the most awful place in Phuket – Bangla road. Maybe it is one of the worst places on Earth, but the aliens will understand perfectly who they are dealing with.
  • I’ve been planning to visit Japan for many years now. When I lived in Moscow, it was far in terms of distance, but now that I live very close, I still have some excuse all the time.
  • Among my recently read books, I can highlight a non-fiction book about negotiations “Always Say No,” and a fiction collection called “Foundation” by Isaac Asimov.
  • The three tracks I would put on a corporate party:
    • Korol’ i Shut (The King and the Jester, Russian horror punk band) – Hozain lesa;
    • Agoria – You’re Not Alone ft. Blasé;
    • FNAF – I hope you die in a fire.

Further plans

  • After 10 years, I see myself as a boring old schoolboy.