Managing team of Cloud Operations, System Operations and QA Automation engineers, which are implementing and managing AWS based cloud platform for financial service like Bank as a Service My responsibility is mix of people management as Team Leader and mentor, Cloud Architect as designer of new platform and prepare ways to migrate from legacy to modern systems and just as an engineer who are missing hands-on job and trying to write code as much as possible Me and my team is working to deliver Java/Scala based micro services to EKS Kubernetes clusters in multi-account hyper-growth environment where priorities shift quickly. Our responsibility includes, but not limited to creating, automating, deploying, operating, and scaling always-on distributed systems. Technology stack: - AWS - Kubernetes & HELM - Terraform - MongoDB, Oracle, Kafka - Gitlab CI - Go (Terratest) - Some kind of legacy based on Chef, Python, Ruby and Jenkins

I built and managed IT/DevOps team as Team Lead and System Architect with reporting directly to company owners I implemented Kubernetes cluster for utility software and designed new architecture of high load REST based game servers, which provides scalability, fault tolerance of all components. It also cloud native with micro services architecture, instead monolithic application was before here Also I moved infrastructure to IoC based on Puppet, improved monitoring and secured all infrastructure by building private VPN network between offices, data centers and clouds Technologies stack: - ELK, Graphite, Grafana, Jenkins, LXD, php, python, nodejs, nginx, Zabbix - ClickHouse, Kafka - Atlassian, Bitwarden, G-Suite, security tools - High load REST applications and mobile applications on client side

Joined Playtika as BigData Engineer and after 6 months was promoted to BigData DBA and on top of that I'm technical lead or mentor my teammates. Our team manages all Big Data related infrastructure, that include several production environments as well as few QA and test environments. This is mix of hardware, VMs and and cloud instances Achievements: - Implemented configuration automation and IoC philosophy based on HA Puppet 5.x cluster with failover to real continuous delivery infrastructure changes to environments. Before I joined Playtika there was no configuration management for Big Data) - Developed Puppet manifests and modules to automate deployment for Apache Storm, Confluent schema registry (with HAProxy & VRRP based failover), Vertica Kafka connector and other tools and services, that not covered by CDH Cloudera Manager - Implemented clustered ELK to collect production logs, configured and deployed by Puppet with automatic failover - No downtime migration all ETL pipelines from one datacenter to another (that was done few times for several projects) - Migrated all Vertica clusters from 7.x to 8.x versions - Together with development team migrated legacy Java and files based solution to modern ETL based on Kafka, Storm and Spark Technologies stack: - CDH 5.x based Hadoop, Kafka, Spark, Hive, Hue etc - HP Vertica analytic database and utils like Kafka Vertica connector - Docker, ELK, Graphite, Grafana, Jenkins, Nexus etc - Python self developed tools for proactive monitoring and automation

Joined as DevOps architect to Melsoft games, which is a subsidiary of Wargaming.net pointed on casual mobile games development. After few months and team growing I was promoted to Head of IT The main objectives for me and my team, is the implementation DevOps philosophy, as well as practical implementation of continuous integration, testing and code / product delivery. In addition to ordinary tasks, such as management and improving of existing infrastructure, I'm producing software development processes that follow the principles of CICD (CITD), as well as the trend of "management of infrastructure and environments in the code" Achievements: - Built team and processes - Implemented Puppet as IoC and configuration management (company used only Fabric before to deploy automation and nothing to configuration management) - Together with R&D team released My Cafe game - Designed and implemented design of new game - Tile Tactics - Fixed teams issues with PostgreSQL performance related to bad application and DB structure design - Implemented security solutions based on Cisco firewalls and routers on both sides - office and data centers, that allowed people work transparent and secure

I'm working as Senior System Engineer in large international team that provides tools for SEO-analysts, research and assessment of competitors' sites and other tools, related to SEO and Internet advertising. From everyday tasks is the support of more than 17 development teams, each of which creates a micro-service, which subsequently integrated into the main service and become part of the semrush.com More than 600 servers, - physical, virtual (Proxmox kvm, openvz), cloud (Amazon, OpenStack) Also my position involves solving infrastructural and architectural tasks and continuous improvement of service Current achievements in this position: - Design and implementation the protected private network in the datacenter (DMZ), based on cluster of two Cisco ASA-5585-S20, which are operating in active-standby mode. As result the creation a single point of access control to project infrastructure, include servers and services. Also this solution has excluded the human factor, when you forget to configure the firewall or the firewall was configured incorrectly by mistake. The solution is continuously evolving and improving by me. - Implemented Open Stack distribution and services deployment full automation by Puppet from scratch - Automation through the puppet to deploy and manage PostgreSQL database server in master-slave replication. This solution also allows automatically restore slave servers from backup in cases of broken replication. - Design and implementation of distributed NoSQL key-value storage, based on Redis and Sentinel, as well as automation deployment and management of this solution through the Puppet (writing modules and roles) And much more

Management of ETL and CICD processes in BigData applications based on Apache Hadoop (Cloudera distributive). Three clusters, two of which are used in the production mode using services such as HBase, YARN, Oozie, Hue, Hive, Spark, Solr and Phoenix Total capacity of BigData clusters is more than 4Pb CICD process built on Jenkins, and the distribution code on git and nexus. Service scripts based on bash and python Achievements: - Migrated several clusters from CDH 4.x to 5.x - Implemented Chef based automation

Research and deployment HP Vertica database Development of related utilities for automatically load data from MySQL to HP Vertica, as such as monitoring tools (http2tcp wrapper as daemon on python, munin and nagios plugins). You can see these utilities in my github repository https://github.com/hostadmin/tools Also, regarding HP Vertica, I have researched and deployed data model, created and optimized projections with first on Community Edition (max 3 nodes) and also in production on Enterprise Edition with two clusters - 6 nodes on main cluster and 4 nodes on failover cluster, both with Management Console and Haproxy + Keepalived VRRP + Self-developed python daemon as health-checker to provide full connectivity and high availability, cause it deployed as urgent service for business model with 24/7 availability without exceptions Also I have built and optimized FreeDBS kernel, sysctl/bootloader configs and tuning nginx on web-balancers, resulting in increased productivity of the web-cluster, and to decrease latency on provide dynamic content to users Of course, half part of work is support main service of this company - openx based, self-developed AdServer with summary traffic about 5-6 Gbit/s and 300k rps, such as support of related services like stats, erp, parthner portal and etc. I have optimized MySQL and deployed distributed structure of MySQL master and slave servers to provide failover and HA Start work to automatically deploy servers via Ansible (developed for web-cluster, but not yet for another parts of system) Deployed Percona Server and Galera based multi-master MySQL cluster with balancing and failover as replace of main MySQL master server Because FreeBSD 9.x have some issue with performance on multi-CPU systems I have migrated web-cluster and few another services to Oracle Linux 6.5 x64 with Unbreakable Enterprise kernel 3.x As Team Lead I deployed KPI, Agile-style weekly sprints and provided a friendly and professional atmosphere in the team of five system administrators, allowing the department to do the job effectively and increase the availability of services.

Support hosting companies and hi-load services (like turbobit.net for example) with 2 hours, half hour and 5 minutes SLA. Live support via livechat, skype, icq, email and ticket board Working with high stress environment with a lot parallel tasks to manage production incidents of 200+ customers and 1600+ servers and VMs with different products, software and hardware