Anton Novikov — DevOps Engineer
Location: Yerevan, Armenia · Open to EU remote or relocation
Email: me@antonnovikov.com · Web: antonnovikov.com
LinkedIn: linkedin.com/in/b-iii-rh-minus · GitHub: github.com/biiirhminus
Summary
DevOps engineer with 8 years in infrastructure — from Linux sysadmin to running cloud platforms end-to-end. I build GitOps platforms, CI/CD pipelines, observability stacks, IaC, and internal tools. Enterprise: GDS at airline scale, 500+ VMs, mission-critical 24/7. Startup: sole DevOps, built from scratch. Now managing 6 Kubernetes clusters from one git repo with FluxCD v2.
I write my own tools — FastAPI dashboards, Alertmanager integrations, CLI tools — and treat them as production code: versioned, containerised, deployed with Helm.
Looking for a Middle / Middle+ DevOps or Platform Engineer role in an EU company.
Skills
| Orchestration | Kubernetes (k3s, Helm, Kustomize, FluxCD v2), Docker Swarm, Docker Compose |
| GitOps | FluxCD v2 (hub-and-spoke multi-cluster), Kustomize, SealedSecrets |
| IaC | Terraform, Ansible |
| CI/CD | GitLab CI/CD (self-hosted & SaaS), GitHub Actions |
| Cloud | Yandex Cloud (compute, managed k8s, S3, DNS, CDN, IAM) |
| Observability | Prometheus, Grafana, Victoria Logs, ELK, OpenTelemetry, Zabbix, Sentry |
| Networking | Cilium, APISIX, Nginx, HAProxy, WireGuard |
| Data | Kafka, RabbitMQ, ClickHouse, Cassandra, MongoDB, Redis, MinIO, PostgreSQL, MySQL |
| Security | HashiCorp Vault, Trivy Operator, SealedSecrets, cert-manager, RBAC |
| Languages | Python, FastAPI, Bash, JavaScript |
Experience
DevOps Engineer · Sirena-Travel group
Aug 2022 – Present · Hybrid → Remote
Worked across three products at Sirena-Travel — from the parent GDS to newer divisions. I ran the full infrastructure at each step.
Red Rose Traveltech — EU B2B subsidiary · 2026 – Present
- Built the full platform from an empty Yandex Cloud account — VPC, managed PostgreSQL/Redis, S3, DNS, load balancers, all in Terraform. No manual console changes
- GitLab CI/CD per service: test → build → Helm deploy → promotion gates (dev → staging → prod) with auto-rollback — commit to prod in under 5 min
- Prometheus + Grafana wired to SLO targets, alerts through Alertmanager, TLS from cert-manager + Let's Encrypt — 99.9%+ on the booking APIs
- Automated recurring ops with n8n — partner onboarding, SLA alerts, Terraform drift — no manual work needed
MixVel — agent GDS division · 2025 – Present
- Designed and run a FluxCD v2 hub-and-spoke platform: one hub reconciles five spokes, secrets sealed in git — no kubectl apply in prod
- 3-layer Kustomize (base / overlays / per-env patches) keeps 90%+ of manifests shared with zero drift; k3s + Cilium on-prem, Yandex Managed Kubernetes in cloud, APISIX at the edge
- Ansible bootstraps a new cluster end to end (k3s, Flux, Cilium, node tuning) in under 30 min; the whole data layer runs as FluxCD HelmReleases — Kafka, RabbitMQ, Cassandra, ClickHouse, MongoDB, Redis, MinIO
- kube-prometheus-stack, Victoria Logs, ELK, OpenTelemetry for observability; Vault + Trivy Operator for security; ~20% less compute after rightsizing requests/limits
- Wrote the tooling the team uses daily: env-view (live cluster health), abot (Alertmanager webhook router), trivy-to-sonarqube, confluence-publisher — all Helm-packaged
Sirena-Travel — parent GDS · 2022 – present
- Moved a 24/7 GDS from bare VMs to containers — XEN/libvirt → Docker Swarm → Kubernetes across 500+ nodes — +40% utilisation, no downtime
- Built GitLab CI/CD from scratch for a polyglot estate (Java, Python, PHP, .NET, C++): tests, SonarQube, Trivy, build, Helm deploy — deploy from hours to under 10 min
- Promotion gates + auto-rollback cut bad-deploy incidents by 65%; Prometheus across 500+ nodes with 30+ dashboards, on-call with runbooks — MTTR 120 → 20 min
- Config as code with Ansible (zero drift); zero-downtime quarterly OS patching across the fleet
Stack: Kubernetes · FluxCD v2 · Kustomize · Helm · Cilium · APISIX · Terraform · Ansible · GitLab CI/CD · Docker Swarm · Prometheus · Grafana · Victoria Logs · ELK · OpenTelemetry · Kafka · RabbitMQ · Cassandra · ClickHouse · Vault · SealedSecrets · Trivy · PostgreSQL · n8n
Middle Sysadmin / Junior DevOps Engineer · ADV/web-engineering
Feb 2020 – Aug 2022 · 2 yrs 7 mos · Hybrid
Hybrid infrastructure for web applications — on-premise and cloud.
- Virtualisation on Proxmox and VMware ESXi; Docker CE/Compose; first Kubernetes + Helm
- Linux VMs (Nginx, PHP-FPM, HAProxy, MySQL/MariaDB) and web stacks (Bitrix, Symfony, Vue, React); Active Directory account administration
- CI/CD with GitLab CI and Jenkins; Ansible config management; monitoring with Zabbix, Grafana, Prometheus
- Bash/Python automation; C# (Console/WPF) tooling for MS Office; remote management of MikroTik/HP/DELL/SuperMicro hardware; user support via Jira/Confluence
Stack: Proxmox · VMware ESXi · Docker · Kubernetes · Helm · Ansible · GitLab CI · Jenkins · Nginx · HAProxy · MySQL · Zabbix · Grafana · Prometheus · Bash · Python
Junior / Middle Sysadmin · TenderTech
Nov 2018 – Feb 2020 · 1 yr 4 mos · On-site
Where it started — system administration and user support.
- User support (Windows/Linux, MS Office, 1C); account administration across AD, Atlassian, GitLab
- Proxmox virtualisation; Linux VMs with Nginx/PHP-FPM, Git
- Office IT: PC repair, hardware procurement, inventory, AXIS video surveillance, UniFi
- Server hardware: SuperMicro, DELL, MikroTik
Stack: Proxmox · Linux · Nginx · PHP-FPM · Git · Active Directory · Atlassian · MikroTik · SuperMicro · DELL
Contract & Freelance
DevOps Engineer · MTS (contract)
2026 · Remote
Crypto / fintech division — blockchain settlement platform (VED) and B2C exchanges (neobridge.ru / cloud-crypto.ru).
- Infrastructure on MWS (MTS Web Services) via Terraform with the custom
mws-cloud/mwsprovider — 100% IaC - kubeadm Kubernetes, 8 environments (dev/stage/test/prod × VED + B2C); node taints isolating blockchain workloads; ArgoCD GitOps — zero manual kubectl in prod
- Reusable GitLab CI job library: Kaniko builds, ArgoCD Helm deploy, Trivy + custom OPA policy, SonarQube, Semgrep; Jsonnet pipeline generation
- Ansible roles for the full cluster lifecycle; Kafka via Strimzi, HashiCorp Vault for secrets, WireGuard for cluster access
Stack: Kubernetes · kubeadm · ArgoCD · Terraform · Ansible · GitLab CI · Kaniko · Calico · Longhorn · Strimzi · Kafka · Vault · Prometheus · Grafana · cert-manager · Trivy · Semgrep · MWS
DevOps Engineer · Flowerave (contract)
2024 – 2025 · Remote
Ticketing startup. Sole DevOps engineer — built the entire infrastructure from scratch.
- Full cloud infra on Yandex Cloud via Terraform — VPCs, managed PostgreSQL, Redis, S3, DNS. 100% IaC; env provisioning 3 days → 30 min
- Production Kubernetes + self-hosted GitLab CE: tests, Docker builds, container registry, zero-downtime Helm deploys
- Observability: Sentry, Prometheus + Grafana, ELK. PostgreSQL backups to S3 with WAL archiving — RPO < 1 hr, RTO < 30 min. Zero data loss in 15 months
Stack: Yandex Cloud · Terraform · Kubernetes · Helm · GitLab CI/CD · Sentry · Prometheus · Grafana · ELK · PostgreSQL · Redis · Python · Nginx · Docker
Languages
- Russian — Native
- English — Upper-intermediate (B2)
Open to
DevOps Engineer (Middle / Middle+) · Platform Engineer · SRE
EU remote or relocation — Netherlands, Germany, Czech Republic, Portugal, Poland, Austria