
Mikalai Saroka
Senior DevOps Engineer, MLOps
DevOps professional with over 8 years in IT and 5+ years specializing in DevOps. Skilled in designing and managing scalable Kubernetes clusters on AWS, with proven expertise in MLOps, AWS cloud infrastructure, and infrastructure-as-code using Terraform to build reliable and secure systems. Adept at automating workflows to streamline CI/CD pipelines and boost operational efficiency. A collaborative team player with a passion for continuous learning and staying ahead in the fast-evolving tech landscape.
Skills
Kubernetes
Master
- ingress-nginx
- karpenter
- helm
- external-dns
Amazon Web Services (AWS)
Master
- EKS
- EC2
- S3
- IAM
- VPC
- RDS
- Route53
- MSK
- Sagemaker
CI/CD
Master
- Azure DevOps
- Gitlab
- Github Actions
- Jenkins
- ArgoCD
Tools
Master
- Terraform
- Ansible
- Docker
MLOps
Advanced
- MLFlow
- DVC
- Airflow
- Databricks
- JupyterHub
- Models training and deployment
- LLM
Monitoring & Observability
Master
- Prometheus
- Victoria Metrics
- Grafana
- Loki
- Tempo
- OpenTelemetry
- Alertmanager
Programming and Scripting
Advanced
- C#
- .NET Framework and Core
- PowerShell
- Python
- Bash
Team Leadership
Advanced
Work Experience (4)
Apr 2024 - Current
Senior DevOps/MLOps Engineer
Social Discovery Group
- AWS
- EKS
- Terraform
- Helm
- Github
- Gitlab
- Grafana
- Victoria Metrics
- OpenTelemetry
- Loki
- Tempo
- Alertmanager
- MLFlow
- DVC
- Airflow
- Databricks
- JupyterHub
- S3
- RDS
- MSK
- Route53
- Sagemaker
- Postgresql
Worked as a Senior DevOps/MLOps on 2 core products, responsible for infrastructure on AWS, all CI/CD pipelines on Gitlab and Github, MLOps processes including models learning and deployment.
- Designed and deployed project infrastructure on AWS
- Deployed and maintained on-prem Gitlab
- Created complex pipelines on Gitlab and Github
- Created Helm charts for all project and dependency services
- Automated deployment of infrastructure resources on AWS using Terraform
- Designed and implemented model learning process on AWS Sagemaker and EKS cluster
- Automated model deployment process on AWS EKS
- Deployed and maintained observability platform based on Victoria Metrics, Grafana, Loki, Tempo
Mar 2022 - Apr 2024
Senior DevOps Engineer
EPAM Systems
- AWS
- EKS
- Terraform
- Helm
- Azure DevOps
- ArgoCD
- Grafana
- Victoria Metrics
- OpenTelemetry
- Loki
- Tempo
- MLFlow
- DVC
- S3
- Route53
- Postgresql
- RabbitMQ
- Tanzu Kubernetes Grid
- Automated build and deployment process using Azure DevOps
- Automated deployment process for Tanzu Kubernetes Grid cluster
- Performed migration from on-prem Kubernetes cluster to the Tanzu Kubernetes Grid
- Created Helm charts for project resources and services
- Automated deployment of required infrastructure resources on AWS using Terraform
- Performed migration from Tanzu Kubernetes Grid to AWS EKS
- Designed and implemented deployment of services using ArgoCD
Jul 2019 - Mar 2022
Teamlead DevOps Engineer
Coherent Solutions
- Maria DB
- MS SQL Server
- Ansible
- Jenkins
- Terraform
- Python
- Bash
- Gerrit
- Maven
- Docker
- Grafana
- Prometheus
- Selenoid
- Vagrant
- Envoy
- Nexus
- SonarQube
- Keycloak
- NiFi
- AWS
- Rackspace
- Managed team of 5 DevOps members (task planning, team coaching, etc.)
- Automated production deployment process, reducing deployment to production from 2 days to 3 hours
- Performed production deployments and maintenance
- Refactored Jenkins pipelines to declarative approach
- Automated various manual tasks using python and ansible
- Performed migration from Rackspace to AWS
- Designed and implemented monitoring for the project
Jan 2016 - Jul 2019
QA Automation Engineer
Coherent Solutions
Education (2)
2018 - 2020
Master's Degree in
Engineer Psychology And Ergonomics
Belarusian State University of Informatics and Radioelectronics
2012 - 2017
Bachelor's Degree in
Information Systems and Technologies
Belarusian National Technical University
Certificates
CKA: Certified Kubernetes Administrator
The Linux Foundation
Languages
English
Fluent
Russian
Native Speaker