June 8, 2017 at 16:26Interview with Yakov Shuvaev about a team of engineers, motivation and interviews
One of the features of the work of a large integrator is that everyone knows or roughly knows what exactly this integrator does and for which clients. The latter is usually brought to the attention of the masses either by discreet press releases, or by the integrator logo in the list of customer partners. But very often many do not know how exactly such work is carried out, and forget that living people work for the large company logo.
We will try to change the situation a bit - we offer you an interview with the head of one of the technical departments of LANIT, Yakov Shuvaev. The department specializes in implementing solutions in the field of building a reliable and secure software infrastructure for information systems, services that provide the production development cycle, as well as everything related to databases and information security. Jacob, tell us a little about yourself and briefly about your department. What are you doing and what are you working on now?
I'll probably start with my studies. I am from RUDN. He studied at the Physics and Mathematics specialty "Applied Mathematics and Computer Science". The first three courses, the study took all the time, the fourth appeared a bit of freedom, and I decided that it was time to get to work. It was 2006. There were three options. The main plan was to go to those companies about which there were real reviews from my friends. There were two such options, and LANIT was the first on the list. My classmate, who studied in parallel flow for some time already worked there. He arranged for me my first ever interview at an IT company. At this interview, I was met by my immediate future manager, head of the development team and vice president of the company, who was not vice president then, but was a simple project manager - they were in the same box in the open space. At that time, I had little idea of what roles there are in IT companies, so I went to the same position as my classmate - “Trainee Developer”. By the way, if I wanted to do what I’m doing now, it wouldn’t work out for me - at that time there were a couple of database experts in the structure of the unit, and there weren’t any admins or security engineers. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. as my classmate - "Developer-trainee." By the way, if I wanted to do what I’m doing now, it wouldn’t work out for me - at that time there were a couple of database experts in the structure of the unit, and there weren’t any admins or security engineers. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. as my classmate - "Developer-trainee." By the way, if I wanted to do what I’m doing now, it wouldn’t work out for me - at that time there were a couple of database experts in the structure of the unit, and there weren’t any admins or security engineers. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. it wouldn’t work out for me - at that time there were a couple of database specialists in the structure of the unit, and there weren’t any admins or security engineers at all. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. it wouldn’t work out for me - at that time there were a couple of database specialists in the structure of the unit, and there weren’t any admins or security engineers at all. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks.
What other options were there? You said there were three.
If I chose the second option, now I would probably work for one of the companies acquired by Rostelecom. I think that is also a good option.
The third option, a backup, was to post a resume on a job site, wait for feedback, and send letters to potential employers. Fortunately, the second and third options were not useful.
At one of the first projects, I was instructed to deploy the system at one of the customers. I liked to assemble a working system from a set of parts. When it was necessary to assemble a cluster from application servers, then still BEA Weblogic, a "new marvelous world" opened.
I found out that in order for the system to work quickly, reliably and safely, one code is not enough. Plus, as practice has shown, if the code works with the developer, this does not mean that you can just take it and run it in a cluster.
Further more. There were new projects on which I completely switched from development to administration. So in the projects of the division in which I worked, there appeared Linux, SVN and the continuous assembly server in its current form. The only thing - it confused me that for a long time I could not find on the Internet a suitable description of my IT role in the project and team. It was a little wider than system administration, the term DevOps appeared only in 2008, and I did not stumble upon the abbreviation SRE then. As a result, I puzzled for a long time over what I was actually doing. It bothered.
With these thoughts, I wrote to my supervisor, after which he proposed to call this activity “IT infrastructure”, and accordingly my role on projects as “IT infrastructure specialist”. Life made sense and rushed ...
Yes, and as I already understand, at that time I was engaged in infrastructure better than I wrote the code.
In 2009, a project was launched to develop one of the state portals. The production process began to change dramatically in accordance with the flow and scale of new tasks: the team grew, the number of test stands grew. The stands had to be created, maintained up to date, it was necessary to prepare for the launch of the system in commercial operation. I began to miss one on all activities, and we began to expand the staff of "infrastructure managers". As I already said, at that time there were Oracle database specialists in the staff, which it was not entirely clear which administrative unit to relate to: either development or administration. In the end, they decided that I would oversee them.
At the time the sector was established, 5 people worked in it, including me.
By 2016, our division opened additional regional offices in Chelyabinsk, Perm, Saratov and Izhevsk. “The dog has grown a bit”, and in 2016 our team became distributed - over 30 people. The question arose about its transformation into a larger administrative unit of the unit — the department.
In preparation for this, it was necessary to solve a very "difficult" question - what should be called the department. On the one hand, the team has long been known for such keywords as “infrastructure” and “databases”. On the other hand, the area of our activity has expanded significantly since 2010, and the word "infrastructure" is associated with many people with hardware, network, data centers - everything that other LANIT units professionally did and does.
At the interviews that I conduct, telling what our department does, I usually draw a picture in the form of a pyramid, which conditionally represents the levels of the abstract information system, starting with the iron on which everything works, and ending with the application software that our unit develops. The upper level “Application software” is the level that the development department is engaged in. It implements the application functionality that is needed by the customer of the information system.
Below are the levels that our department deals with. We close some levels of the pyramid completely, some partially. By partial closure, I understand that the employees of the department should at least understand what is happening there and there, in order to solve the main problem - to make the software work quickly, reliably and safely.
These tasks can be solved by building an “infrastructure perimeter” around the application code, but it’s obvious that this is most often impossible, so we work closely with the development team and the team that works with hardware, network and virtualization to achieve these goals.
It should be understood that there is no “fence” with the software level in the upper part of the pyramid and the level of physics below. At a minimum, in order to answer the question of whether it is necessary to optimize the code, add hardware or “tighten” the OS, one must either understand all these levels, or be able to correctly diagnose and formulate a question for the one who understands.
The structure of the department is divided into sectors. The first sector provides services for the operation of information systems. Usually this is the third level of support, which means that the team is aware of what the system is, how it works, what tasks it performs and is directly involved in the system development process. The objectives of this team are to ensure reliable, fast and safe operation of software in the industrial environment and to ensure the continuity of production processes. The main tasks that the sector team solves include: creating and configuring deployment automation solutions, infrastructure and business monitoring, configuration management, cloud resource management, reliability, and everything related to it. By 2017, unlike the time when it all began,
What else did the department do?
The second direction is services. The Jira and Wiki systems have long become an integral part of the production process of virtually any project unit. In addition, on the basis of Jira, in addition to the design, many services of the department are automated, starting from ordering the office and receiving VPN access for remote work and ending with the process of hiring new employees.
The third area of work for the sector is the provision of cloud computing and data storage resources to the project teams of the unit. We provide services for organizing the rental of resources from various cloud providers, such as Amazon, Selectel, OnCloud and automation of the preparation of test environments for projects in the clouds.
The second sector is a team of database experts. At the beginning of its journey, the team specialized only in Oracle DBMS. Over time, the guys gained extensive experience working with the PostgreSQL DBMS, NoSQL RIAK database and a number of products that are somehow related to databases. For example, the same Delphix.
With the expansion of the team's competencies, the class of tasks solved by the team of databases was transformed.
A new class of tasks required the use of new tools?
A source
It's not always the same. Sometimes a task comes, and it needs to be urgently solved. It doesn’t matter what tools. If it turns out that at this moment you can only hold a microscope in your hands, then you will solve all problems, including “driving nails,” with a microscope. Incidentally, this is normal if the customer wants a microscope, pays for a microscope and for a house built with a microscope, and in the end receives what he wanted. From the outside, the situation may look strange, but if the customer is satisfied, then the goal is achieved.
For obvious reasons, we try to prevent such situations. To solve problems, make projects and not look strange at the same time, it is important to be aware of what is happening around you in the industry before receiving a new task or starting a new project. Knowledge of a wide range of technologies and solutions makes it possible to explain to the customer that a microscope is not the best thing in the world, and offer the most optimal option for solving a problem.
And how do you do it?
We try to keep abreast :) We watch webinars, go to conferences, organize meetings. If time permits, we try to use new technologies on some tasks, so that when the time comes, we can offer them.
Clear. Is infrastructure and databases all that you do or are there any other areas?
There is. In 2015, we began to work in the areas of information security and designing disaster-resistant solutions. Among the results can be noted the completion of the design of the reservation scheme of two large state information systems. In one project, the system is reserved for two data centers, in the second - for four.
As far as I know, the same Amazon allows you to do this almost out of the box. What is the difficulty?
Amazon provides this opportunity at the level of its infrastructure or services - depending on what you use. At the same time, depending on the technologies that are used in your project, it will work well only when you take into account the possibility of “moving” between data centers at the code level of your system. Otherwise, the result will be little predictable. In our case, the task was to come up with a backup scheme for a system that has been operating for many years, which would guarantee the set RTO and RPO indicators.
OK. What about security?
In any country there are laws and regulations that require a certain class of systems to comply with a certain set of requirements. In Russia, the main laws in this area are 149, 152 and 63 Federal Laws and the corresponding orders of various departments. Since we are mainly engaged in government information systems, in which personal data can be processed, and an electronic digital signature can also be used, we must take into account the requirements of all this regulation. From the results, I can note the completion of work on certification of one of the information systems that our unit is developing, according to the requirements of the FSTEC, we have certified the software included in this system, and two impact assessments according to the requirements of the FSB.
In addition to activities required by law, inside the unit we train colleagues from other departments, test and develop information security issues in the production and operation of systems.
And what is the name of the department?
"Department of infrastructure solutions, services and data management." Long, but reflects the essence of what we do.
What are you especially proud of while working?
I am proud that I managed to assemble a professional team with which I can solve problems and which I trust.
How do you motivate employees to achieve the desired results? And what is the secret of a successful team, in your opinion?
I’m not specifically motivating anyone. In my work I proceed from two principles. The first - when setting the task, in addition to describing what I am waiting for as a result, I try to outline the context: why and who needs the result of this task. I know from my own experience that there is no stranger state when you try to guess what exactly they want from you and in what form, especially if the task you receive is not from the one who needs the result.
The second - I assume that we have all adults, including interns. To everyone who comes to us, I say that if the task is incomprehensible, you should not try to guess and think out, you need to come up and ask - no one will hit on the forehead, but it can save a lot of time for everyone.
Regarding the secret of success. This question is better to ask my head - the director of the unit :) If he hypothetically answered that our department is successful, I would, again, hypothetically, answered this question with history. Unfortunately, I do not know who the author is, but she is very close to me in spirit. It sounds something like this:
In your opinion, is there a staff shortage in LANIT and in the IT sector as a whole? What kind of specialists does LANIT need today to solve the tasks set for the company? And what needs to be done so that young people remain to work in the company?
Good questions. There are difficulties with hiring specialists, but this situation is typical for the entire market - not only for LANIT. Since the projects that our division is working on do not usually involve work “from here until lunchtime”, the list of specialists who are suitable for us is greatly narrowed.
In my opinion, there are several options for how to be in such a situation.
The first is classic. You can search for the people you need using classic HR approaches for a long time, but the output is always high-quality. Many can argue with me here, but I can say about my experience. I was probably lucky with our HRs.
You can also search for people in the regions, you can take interns and pump them to the tasks you need. All three methods will not make the current team good right now, but in the future they will solve the problem.
The second is to publicize the company on the market, in social networks and attract those who have not even thought about working at LANIT before. Everyone knows about Yandex, Avito, Mail.ru, but what do you know about LANIT? I always ask this question at interviews.
And what do they answer?
Usually, if they come from integrators, then they most likely heard something, but not necessarily about projects that were dealt with by our division. LANIT is a group of companies, and each of them specializes in something. Someone is building cosmodromes, someone is engaged in ATMs - there are different directions. The rest of us either don’t know anything, or they can answer something like “Yes, I heard ... Are you doing public services?” :) And this despite the fact that LANIT is one of the three largest IT companies in Russia for a year , and projects that have been implemented only by our unit are used throughout the country.
The main thing is not to focus on one approach and, fortunately, in this regard we are actively moving in the right direction. We carry out mitaps, went to Habr. Plus, the internship program is gaining momentum, according to which we are looking for young children to our place for training and further employment.
We try to be in trend.
In the trend - this is good. And how to keep employees further?
To begin with, I would like to say that no one is holding anyone by force :) In order for employees not to leave, conditions must be created that would be difficult to part with. First of all, this is the opportunity to do what you love, then, so that employees feel stability and growth prospects in the company, and thirdly, so that they can develop to their own advantage and the company.
To achieve this, firstly, it seems important to me that every employee understands what the unit in which he works is engaged in and lives on. If he does not understand this, he will not understand what and why he is doing and part of what he is.
Obviously, we are not a startup or a product division, although we also have our own software products and internal startups. First of all, we are an integrator who makes projects. Large and complex turnkey projects. What does this mean in practice? This means that we are implementing a full production cycle for the creation of information systems, starting with the collection of requirements and ending with the operation and development of systems already operating in production. The projects we are working on are socially significant and important for the country. One awareness of what you do and how many people use the result of your work motivates very much.
Secondly, each employee must understand what opportunities a job in a company opens up for him, what options he has for development and the possibilities for this.
The unit has accumulated tremendous experience that can be adopted and developed with its help at a frantic speed. I am sure that there are not many companies and divisions that can share the knowledge and competencies that our experts possess. Take and learn!
With theory it’s clear, but how in practice? Does everything work right?
Well ... we are not perfect, but we try to go in this direction. Most of this theory can be “felt” right there in practice. The dynamics of our projects is such that, falling into the cycle of the release of the next release of some federal “spacecraft” at the end of the year, you can get more experience in a month than in another company in a year.
A source
In the beginning, this speed is confusing. Sometimes it’s hard to understand what’s going on. But plunged into this atmosphere, it is already difficult to abandon this rhythm, which motivates you to keep your finger on the pulse of progress and constantly evolve.
And, of course, thirdly - interviews.
In order to keep an employee, one must make sure that he does not need to be kept. And for this it is necessary that people who 100% fit into the team and immediately join the general flow fall into the team. To understand whether this employee will fall into the stream or not, for several hours of interviews - this is another task. Starting communication with candidates, I always first of all synchronize terminology with them. Everyone who comes to us usually worked somewhere before. Everyone has their own understanding of certain terms: someone says that he is an architect, someone talks about the fact that they did highly loaded projects, someone says that they had a lot of data stored in the databases, etc. . Before building a dialogue, you need to understand that you speak the same language. If you watched the movie "Arrival", then here is about the same story.
In the case of our unit, one project can consist of dozens of parts, with each part as a separate large project for other companies. Each such project is done by a separate team consisting of a manager, analyst, development team leader, etc. When at the first meeting a person says that he “participated in 100 projects, supported 1000 servers and deployed more than 9000 databases”, at first glance it seems that a person with such experience should be taken into a team without looking. But, if you start to dive into the details, often everything becomes not so obvious.
Another important thing that I always ask people I’m interviewing is what would they like to do in an IT company if they had a choice, all else being equal? What is the soul, if you want? Since we are constantly opening vacancies for different roles, we, as a rule, do not filter people for a vacancy, but look at people and, if there is such an opportunity, give them a choice - which position should they go to. For me, the main thing is that the person we take will not only do the job well, but also enjoy it - then the likelihood that he will leave will be less.
Fourth, for people to be able to pump, grow, exchange experiences and express themselves, it is necessary to maintain a constant process of accumulating this experience and transferring knowledge. We also have such opportunities.
You probably won't surprise anyone with a corporate knowledge base in the Wiki format now, but it’s worth mentioning about it. Our database contains many pages created by employees, with examples of solving technical problems and articles on various IT topics. These are not Habr format articles or something like that. These are rather notes of employees of varying degrees of detail for their own colleagues, which quickly help beginners to orient themselves in the set of technologies that we use, and quickly plunge into the project or remember something that was a long time ago.
For those employees who are close in the corporate Wiki, and why hide it, for PR too, last year we launched the # TechGuruDay series of events , with Twitter and a channel in Telegram. The rallies are held on the LANIT site with live reports on IT topics. Our employees can speak with invited external speakers. This year we launched a blog on Habré, in which all employees can write. By the way, taking this opportunity, I invite everyone who has an interesting topic for a mitap, write to me on meetup.com - if the topic somehow overlaps with what we are doing, we will organize a joint mitap.
Therefore, in summary, I can say that the reasons for not leaving us, it seems to me, are more than leaving! But, life situations are different, and they also leave us. In these cases, we are usually very sad, but continue to work further. Especially knowing that the departed employees are sad with us and very often after some time come back.
Do you have plans to launch any new activities?
This year I want to hold more TechGuruDay meetings than last year, I want to tell you something interesting at some conference.
What can you wish your colleagues?
Do not stand in one place, constantly develop and move the world forward!
If right now people want to get to your team, where can they find out about your "spacecraft" and who do you need?
Now it will be like the last slide at some presentation from the conference with a large inscription “WE'RE HIRING”.
About the "spacecraft" you can ask questions in the comments. In short, we are looking for specialists in PostgreSQL and Oracle databases, engineers with ansible and python knowledge in the DevOps / SRE team, architects and security engineers. Moreover, all levels. Well, and, like everyone else, we have a site with vacancies , where everything is described in more detail.
We will try to change the situation a bit - we offer you an interview with the head of one of the technical departments of LANIT, Yakov Shuvaev. The department specializes in implementing solutions in the field of building a reliable and secure software infrastructure for information systems, services that provide the production development cycle, as well as everything related to databases and information security. Jacob, tell us a little about yourself and briefly about your department. What are you doing and what are you working on now?
I'll probably start with my studies. I am from RUDN. He studied at the Physics and Mathematics specialty "Applied Mathematics and Computer Science". The first three courses, the study took all the time, the fourth appeared a bit of freedom, and I decided that it was time to get to work. It was 2006. There were three options. The main plan was to go to those companies about which there were real reviews from my friends. There were two such options, and LANIT was the first on the list. My classmate, who studied in parallel flow for some time already worked there. He arranged for me my first ever interview at an IT company. At this interview, I was met by my immediate future manager, head of the development team and vice president of the company, who was not vice president then, but was a simple project manager - they were in the same box in the open space. At that time, I had little idea of what roles there are in IT companies, so I went to the same position as my classmate - “Trainee Developer”. By the way, if I wanted to do what I’m doing now, it wouldn’t work out for me - at that time there were a couple of database experts in the structure of the unit, and there weren’t any admins or security engineers. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. as my classmate - "Developer-trainee." By the way, if I wanted to do what I’m doing now, it wouldn’t work out for me - at that time there were a couple of database experts in the structure of the unit, and there weren’t any admins or security engineers. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. as my classmate - "Developer-trainee." By the way, if I wanted to do what I’m doing now, it wouldn’t work out for me - at that time there were a couple of database experts in the structure of the unit, and there weren’t any admins or security engineers. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. it wouldn’t work out for me - at that time there were a couple of database specialists in the structure of the unit, and there weren’t any admins or security engineers at all. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. it wouldn’t work out for me - at that time there were a couple of database specialists in the structure of the unit, and there weren’t any admins or security engineers at all. After the head of the development team found out that I know little about GWT, JEE and the garbage collector, and my programming experience comes down to writing labs at a university and playing sudoku in C # for PDAs, he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks. he handed me over to my future leader. I was given an A4 sheet with logic tasks. As it turned out later, everyone who was interviewed in the development department went through these tasks.
Interesting fact. Not everyone knows, but the RUDN Physics and Mathematics Department settled firmly in LANIT. Only in one unit in which Yakov works, 5 more graduates of the RUDN Physics Department work.
What other options were there? You said there were three.
If I chose the second option, now I would probably work for one of the companies acquired by Rostelecom. I think that is also a good option.
The third option, a backup, was to post a resume on a job site, wait for feedback, and send letters to potential employers. Fortunately, the second and third options were not useful.
Then
At one of the first projects, I was instructed to deploy the system at one of the customers. I liked to assemble a working system from a set of parts. When it was necessary to assemble a cluster from application servers, then still BEA Weblogic, a "new marvelous world" opened.
I found out that in order for the system to work quickly, reliably and safely, one code is not enough. Plus, as practice has shown, if the code works with the developer, this does not mean that you can just take it and run it in a cluster.
Further more. There were new projects on which I completely switched from development to administration. So in the projects of the division in which I worked, there appeared Linux, SVN and the continuous assembly server in its current form. The only thing - it confused me that for a long time I could not find on the Internet a suitable description of my IT role in the project and team. It was a little wider than system administration, the term DevOps appeared only in 2008, and I did not stumble upon the abbreviation SRE then. As a result, I puzzled for a long time over what I was actually doing. It bothered.
With these thoughts, I wrote to my supervisor, after which he proposed to call this activity “IT infrastructure”, and accordingly my role on projects as “IT infrastructure specialist”. Life made sense and rushed ...
Yes, and as I already understand, at that time I was engaged in infrastructure better than I wrote the code.
DevOps and SRE
In 2009, a project was launched to develop one of the state portals. The production process began to change dramatically in accordance with the flow and scale of new tasks: the team grew, the number of test stands grew. The stands had to be created, maintained up to date, it was necessary to prepare for the launch of the system in commercial operation. I began to miss one on all activities, and we began to expand the staff of "infrastructure managers". As I already said, at that time there were Oracle database specialists in the staff, which it was not entirely clear which administrative unit to relate to: either development or administration. In the end, they decided that I would oversee them.
At the time the sector was established, 5 people worked in it, including me.
By 2016, our division opened additional regional offices in Chelyabinsk, Perm, Saratov and Izhevsk. “The dog has grown a bit”, and in 2016 our team became distributed - over 30 people. The question arose about its transformation into a larger administrative unit of the unit — the department.
In preparation for this, it was necessary to solve a very "difficult" question - what should be called the department. On the one hand, the team has long been known for such keywords as “infrastructure” and “databases”. On the other hand, the area of our activity has expanded significantly since 2010, and the word "infrastructure" is associated with many people with hardware, network, data centers - everything that other LANIT units professionally did and does.
At the interviews that I conduct, telling what our department does, I usually draw a picture in the form of a pyramid, which conditionally represents the levels of the abstract information system, starting with the iron on which everything works, and ending with the application software that our unit develops. The upper level “Application software” is the level that the development department is engaged in. It implements the application functionality that is needed by the customer of the information system.
Below are the levels that our department deals with. We close some levels of the pyramid completely, some partially. By partial closure, I understand that the employees of the department should at least understand what is happening there and there, in order to solve the main problem - to make the software work quickly, reliably and safely.
These tasks can be solved by building an “infrastructure perimeter” around the application code, but it’s obvious that this is most often impossible, so we work closely with the development team and the team that works with hardware, network and virtualization to achieve these goals.
It should be understood that there is no “fence” with the software level in the upper part of the pyramid and the level of physics below. At a minimum, in order to answer the question of whether it is necessary to optimize the code, add hardware or “tighten” the OS, one must either understand all these levels, or be able to correctly diagnose and formulate a question for the one who understands.
The structure of the department is divided into sectors. The first sector provides services for the operation of information systems. Usually this is the third level of support, which means that the team is aware of what the system is, how it works, what tasks it performs and is directly involved in the system development process. The objectives of this team are to ensure reliable, fast and safe operation of software in the industrial environment and to ensure the continuity of production processes. The main tasks that the sector team solves include: creating and configuring deployment automation solutions, infrastructure and business monitoring, configuration management, cloud resource management, reliability, and everything related to it. By 2017, unlike the time when it all began,
Base
What else did the department do?
The second direction is services. The Jira and Wiki systems have long become an integral part of the production process of virtually any project unit. In addition, on the basis of Jira, in addition to the design, many services of the department are automated, starting from ordering the office and receiving VPN access for remote work and ending with the process of hiring new employees.
The third area of work for the sector is the provision of cloud computing and data storage resources to the project teams of the unit. We provide services for organizing the rental of resources from various cloud providers, such as Amazon, Selectel, OnCloud and automation of the preparation of test environments for projects in the clouds.
The second sector is a team of database experts. At the beginning of its journey, the team specialized only in Oracle DBMS. Over time, the guys gained extensive experience working with the PostgreSQL DBMS, NoSQL RIAK database and a number of products that are somehow related to databases. For example, the same Delphix.
With the expansion of the team's competencies, the class of tasks solved by the team of databases was transformed.
A new class of tasks required the use of new tools?
A source
It's not always the same. Sometimes a task comes, and it needs to be urgently solved. It doesn’t matter what tools. If it turns out that at this moment you can only hold a microscope in your hands, then you will solve all problems, including “driving nails,” with a microscope. Incidentally, this is normal if the customer wants a microscope, pays for a microscope and for a house built with a microscope, and in the end receives what he wanted. From the outside, the situation may look strange, but if the customer is satisfied, then the goal is achieved.
For obvious reasons, we try to prevent such situations. To solve problems, make projects and not look strange at the same time, it is important to be aware of what is happening around you in the industry before receiving a new task or starting a new project. Knowledge of a wide range of technologies and solutions makes it possible to explain to the customer that a microscope is not the best thing in the world, and offer the most optimal option for solving a problem.
And how do you do it?
We try to keep abreast :) We watch webinars, go to conferences, organize meetings. If time permits, we try to use new technologies on some tasks, so that when the time comes, we can offer them.
Safety and reliability
Clear. Is infrastructure and databases all that you do or are there any other areas?
There is. In 2015, we began to work in the areas of information security and designing disaster-resistant solutions. Among the results can be noted the completion of the design of the reservation scheme of two large state information systems. In one project, the system is reserved for two data centers, in the second - for four.
As far as I know, the same Amazon allows you to do this almost out of the box. What is the difficulty?
Amazon provides this opportunity at the level of its infrastructure or services - depending on what you use. At the same time, depending on the technologies that are used in your project, it will work well only when you take into account the possibility of “moving” between data centers at the code level of your system. Otherwise, the result will be little predictable. In our case, the task was to come up with a backup scheme for a system that has been operating for many years, which would guarantee the set RTO and RPO indicators.
OK. What about security?
In any country there are laws and regulations that require a certain class of systems to comply with a certain set of requirements. In Russia, the main laws in this area are 149, 152 and 63 Federal Laws and the corresponding orders of various departments. Since we are mainly engaged in government information systems, in which personal data can be processed, and an electronic digital signature can also be used, we must take into account the requirements of all this regulation. From the results, I can note the completion of work on certification of one of the information systems that our unit is developing, according to the requirements of the FSTEC, we have certified the software included in this system, and two impact assessments according to the requirements of the FSB.
In addition to activities required by law, inside the unit we train colleagues from other departments, test and develop information security issues in the production and operation of systems.
And what is the name of the department?
"Department of infrastructure solutions, services and data management." Long, but reflects the essence of what we do.
What are you especially proud of while working?
I am proud that I managed to assemble a professional team with which I can solve problems and which I trust.
About motivation
How do you motivate employees to achieve the desired results? And what is the secret of a successful team, in your opinion?
I’m not specifically motivating anyone. In my work I proceed from two principles. The first - when setting the task, in addition to describing what I am waiting for as a result, I try to outline the context: why and who needs the result of this task. I know from my own experience that there is no stranger state when you try to guess what exactly they want from you and in what form, especially if the task you receive is not from the one who needs the result.
The second - I assume that we have all adults, including interns. To everyone who comes to us, I say that if the task is incomprehensible, you should not try to guess and think out, you need to come up and ask - no one will hit on the forehead, but it can save a lot of time for everyone.
Regarding the secret of success. This question is better to ask my head - the director of the unit :) If he hypothetically answered that our department is successful, I would, again, hypothetically, answered this question with history. Unfortunately, I do not know who the author is, but she is very close to me in spirit. It sounds something like this:
“One leader is interviewed.
- Tell me, here all your employees constantly attend different conferences, receive certificates, go to various trainings and meetings in other companies' offices ... Are you not afraid that they will learn everything from you and leave you?
“I'm afraid that they will not learn anything and will remain.”
About hiring employees
In your opinion, is there a staff shortage in LANIT and in the IT sector as a whole? What kind of specialists does LANIT need today to solve the tasks set for the company? And what needs to be done so that young people remain to work in the company?
Good questions. There are difficulties with hiring specialists, but this situation is typical for the entire market - not only for LANIT. Since the projects that our division is working on do not usually involve work “from here until lunchtime”, the list of specialists who are suitable for us is greatly narrowed.
In my opinion, there are several options for how to be in such a situation.
The first is classic. You can search for the people you need using classic HR approaches for a long time, but the output is always high-quality. Many can argue with me here, but I can say about my experience. I was probably lucky with our HRs.
You can also search for people in the regions, you can take interns and pump them to the tasks you need. All three methods will not make the current team good right now, but in the future they will solve the problem.
The second is to publicize the company on the market, in social networks and attract those who have not even thought about working at LANIT before. Everyone knows about Yandex, Avito, Mail.ru, but what do you know about LANIT? I always ask this question at interviews.
And what do they answer?
Usually, if they come from integrators, then they most likely heard something, but not necessarily about projects that were dealt with by our division. LANIT is a group of companies, and each of them specializes in something. Someone is building cosmodromes, someone is engaged in ATMs - there are different directions. The rest of us either don’t know anything, or they can answer something like “Yes, I heard ... Are you doing public services?” :) And this despite the fact that LANIT is one of the three largest IT companies in Russia for a year , and projects that have been implemented only by our unit are used throughout the country.
The main thing is not to focus on one approach and, fortunately, in this regard we are actively moving in the right direction. We carry out mitaps, went to Habr. Plus, the internship program is gaining momentum, according to which we are looking for young children to our place for training and further employment.
We try to be in trend.
About retention
In the trend - this is good. And how to keep employees further?
To begin with, I would like to say that no one is holding anyone by force :) In order for employees not to leave, conditions must be created that would be difficult to part with. First of all, this is the opportunity to do what you love, then, so that employees feel stability and growth prospects in the company, and thirdly, so that they can develop to their own advantage and the company.
To achieve this, firstly, it seems important to me that every employee understands what the unit in which he works is engaged in and lives on. If he does not understand this, he will not understand what and why he is doing and part of what he is.
Obviously, we are not a startup or a product division, although we also have our own software products and internal startups. First of all, we are an integrator who makes projects. Large and complex turnkey projects. What does this mean in practice? This means that we are implementing a full production cycle for the creation of information systems, starting with the collection of requirements and ending with the operation and development of systems already operating in production. The projects we are working on are socially significant and important for the country. One awareness of what you do and how many people use the result of your work motivates very much.
Secondly, each employee must understand what opportunities a job in a company opens up for him, what options he has for development and the possibilities for this.
The unit has accumulated tremendous experience that can be adopted and developed with its help at a frantic speed. I am sure that there are not many companies and divisions that can share the knowledge and competencies that our experts possess. Take and learn!
With theory it’s clear, but how in practice? Does everything work right?
Well ... we are not perfect, but we try to go in this direction. Most of this theory can be “felt” right there in practice. The dynamics of our projects is such that, falling into the cycle of the release of the next release of some federal “spacecraft” at the end of the year, you can get more experience in a month than in another company in a year.
A source
In the beginning, this speed is confusing. Sometimes it’s hard to understand what’s going on. But plunged into this atmosphere, it is already difficult to abandon this rhythm, which motivates you to keep your finger on the pulse of progress and constantly evolve.
And, of course, thirdly - interviews.
About interviews
In order to keep an employee, one must make sure that he does not need to be kept. And for this it is necessary that people who 100% fit into the team and immediately join the general flow fall into the team. To understand whether this employee will fall into the stream or not, for several hours of interviews - this is another task. Starting communication with candidates, I always first of all synchronize terminology with them. Everyone who comes to us usually worked somewhere before. Everyone has their own understanding of certain terms: someone says that he is an architect, someone talks about the fact that they did highly loaded projects, someone says that they had a lot of data stored in the databases, etc. . Before building a dialogue, you need to understand that you speak the same language. If you watched the movie "Arrival", then here is about the same story.
In the case of our unit, one project can consist of dozens of parts, with each part as a separate large project for other companies. Each such project is done by a separate team consisting of a manager, analyst, development team leader, etc. When at the first meeting a person says that he “participated in 100 projects, supported 1000 servers and deployed more than 9000 databases”, at first glance it seems that a person with such experience should be taken into a team without looking. But, if you start to dive into the details, often everything becomes not so obvious.
Another important thing that I always ask people I’m interviewing is what would they like to do in an IT company if they had a choice, all else being equal? What is the soul, if you want? Since we are constantly opening vacancies for different roles, we, as a rule, do not filter people for a vacancy, but look at people and, if there is such an opportunity, give them a choice - which position should they go to. For me, the main thing is that the person we take will not only do the job well, but also enjoy it - then the likelihood that he will leave will be less.
About Mentoring
Fourth, for people to be able to pump, grow, exchange experiences and express themselves, it is necessary to maintain a constant process of accumulating this experience and transferring knowledge. We also have such opportunities.
You probably won't surprise anyone with a corporate knowledge base in the Wiki format now, but it’s worth mentioning about it. Our database contains many pages created by employees, with examples of solving technical problems and articles on various IT topics. These are not Habr format articles or something like that. These are rather notes of employees of varying degrees of detail for their own colleagues, which quickly help beginners to orient themselves in the set of technologies that we use, and quickly plunge into the project or remember something that was a long time ago.
About mitaps
For those employees who are close in the corporate Wiki, and why hide it, for PR too, last year we launched the # TechGuruDay series of events , with Twitter and a channel in Telegram. The rallies are held on the LANIT site with live reports on IT topics. Our employees can speak with invited external speakers. This year we launched a blog on Habré, in which all employees can write. By the way, taking this opportunity, I invite everyone who has an interesting topic for a mitap, write to me on meetup.com - if the topic somehow overlaps with what we are doing, we will organize a joint mitap.
Therefore, in summary, I can say that the reasons for not leaving us, it seems to me, are more than leaving! But, life situations are different, and they also leave us. In these cases, we are usually very sad, but continue to work further. Especially knowing that the departed employees are sad with us and very often after some time come back.
About plans for the future
Do you have plans to launch any new activities?
This year I want to hold more TechGuruDay meetings than last year, I want to tell you something interesting at some conference.
What can you wish your colleagues?
Do not stand in one place, constantly develop and move the world forward!
If right now people want to get to your team, where can they find out about your "spacecraft" and who do you need?
Now it will be like the last slide at some presentation from the conference with a large inscription “WE'RE HIRING”.
About the "spacecraft" you can ask questions in the comments. In short, we are looking for specialists in PostgreSQL and Oracle databases, engineers with ansible and python knowledge in the DevOps / SRE team, architects and security engineers. Moreover, all levels. Well, and, like everyone else, we have a site with vacancies , where everything is described in more detail.