Senior DevOps Engineer | System Architecture Portfolio

// Engineering Philosophy

I don't just write scripts; I engineer resilience. My evolution from SysAdmin to DevOps and now towards Technical Leadership is driven by a single obsession: how do we scale complexity without scaling chaos?

Specializing in the Kafka ecosystem and Kubernetes orchestration, I treat infrastructure as a product. If a deployment creates friction, it's a bug. If a schema change breaks a consumer, it's an architectural failure.

My approach combines the precision of a kernel developer with the strategic foresight of a system architect. I build platforms that empower developers to ship faster, sleep better, and innovate fearlessly.

user@moscow-node-01:~

$ whoami

Senior DevOps Engineer / Lead Candidate

$ cat career_goals.json

{
  "current_focus": "Distributed Systems",
  "next_step": "Technical Leadership",
  "values": ["Automation", "Mentorship", "Reliability"]
}

System Arsenal //

01. Data & Streaming

Apache Kafka

Debezium CDC, Schema Registry, MirrorMaker 2, KSQL.

PostgreSQL

HA configurations, Patroni, PGBouncer, tuning.

02. Orchestration

Kubernetes

Operator pattern, CRDs, Network Policies, Istio.

Helm

Library charts, complex dependency management.

03. Automation

GitLab CI

Dynamic pipelines, runners scaling, compliance.

Ansible

Infrastructure provisioning, config drift mgmt.

Mission Logs CONFIDENTIAL

INCIDENT_ID: KFK-MIG-01

Zero-Downtime Kafka Cluster Migration

The Challenge

Migrating a production Kafka cluster handling 50k msgs/sec to new hardware without data loss or consumer lag spikes.

The Solution

Implemented MirrorMaker 2 for active-active replication. Orchestrated a blue-green switchover using DNS weighting and staggered consumer group restarts.

> uptime check
100% available during migration
> lag check
Max lag: 450ms (within SLA)
> status
MIGRATION SUCCESSFUL

INCIDENT_ID: ART-REP-99

Global Artifact Repository Unification

Consolidated 5 disparate Nexus instances into a centralized Artifactory HA cluster. Wrote custom Python scripts to validate checksums for 10TB of artifacts during transfer. Implemented aggressive caching policies for remote teams, reducing build times by 40%.

Python Artifactory S3

INCIDENT_ID: CI-SCALE-04

Scaling GitLab CI for Monorepo

Reduced pipeline duration from 45m to 12m for a monolithic codebase. Used DAG (Directed Acyclic Graph) for stage parallelism, implemented distributed caching with MinIO, and built custom autoscaling runners on spot instances to reduce costs by 60%.

NEXT_LEVEL: TECH LEAD

Transitioning from individual contributor to force multiplier.

Strategy

Aligning technical debt payoff with business velocity. Choosing boring technology for critical paths.

Mentorship

Elevating junior engineers through code reviews, RFC processes, and blame-free post-mortems.

Visibility

Implementing SLOs/SLIs that matter. Making system health visible to stakeholders, not just ops.

OPEN TO OPPORTUNITIES

Initialize Handshake

Moscow based. Remote ready. Global mindset.

Designing Reliable Systems At Scale