< ALEKSANDR GRIBAKIN />

Lead Full Stack & DevOps Engineer
15 Years in Code
Fifteen years building and scaling distributed systems, across the industry's move from bare-metal servers and monoliths to cloud-native, Kubernetes and GitOps. I take on the work teams postpone: paying down technical debt, building deployment from nothing, and keeping systems running.
Kubernetes Terraform Docker GitLab CI PostgreSQL Go Node.js

The Work in Numbers

Career Timeline

Reference Architectures

How I approach these kinds of problems — the method, not a client list

Migration Bare-Metal to Hybrid Cloud Terraform, Ansible, AWS, GCP, Linux △ 30-40% cost cut / staged cutover Moving off servers you own and pay for whether or not they are busy. Done in stages — take stock, match the new setup to the old, then move traffic — so the business never rides on one big switchover night. Platform Kubernetes Platform Kubernetes, Helm, ArgoCD, Terraform, Prometheus △ 99.9% uptime / self-serve namespaces Shared infrastructure your developers can use without first becoming infrastructure specialists. They ask for space and get it already set up, so new work starts the same day instead of waiting on someone else. IaC Infrastructure as Code Baseline Terraform, Ansible, GitLab CI, Vault, Linux △ 100% reproducible / 3 environments Every server described in a file, so any of them can be rebuilt from nothing in minutes. Nothing depends on one machine nobody dares touch, or on what somebody changed by hand two years ago and did not write down. Tooling Local and Production Parity Docker, Docker Compose, Terraform, Make, Linux △ One image / 4 environments A developer’s laptop, the test system and the live one all run the same thing. That removes an entire category of delay: bugs that appear only after release, which nobody can reproduce and everybody has to guess at. Delivery Standardised Delivery Pipelines GitLab CI, Jenkins, Docker, Helm, GitHub Actions △ 15 min release / dozens of services One release process shared by every part of the system, instead of each team maintaining its own. Shipping a change goes from a planned multi-day operation to about fifteen minutes. Deployment Zero-Downtime Deployment Kubernetes, ArgoCD, NGINX, Helm, Prometheus △ Zero downtime / 90s rollback Updates go out without taking the service down, and a release that starts misbehaving is pulled back automatically before most customers see anything. No more maintenance windows announced in advance. GitOps GitOps Delivery Model ArgoCD, Git, Helm, Kustomize, Kubernetes △ Git as source of truth / auto-reconcile One written record of how the system is supposed to be set up, and the system corrects itself back to it. Changes made by hand at three in the morning do not survive to become next month’s mystery. Architecture Monolith to Services PHP, Symfony, Go, gRPC, RabbitMQ △ Incremental cutover / no freeze Splitting one large aging system into parts that can be worked on separately — while it stays open for business the whole time. No freeze, and no rewrite-from-scratch project that runs two years and delivers nothing. API API Layer and Contracts Node.js, TypeScript, OpenAPI, gRPC, NGINX △ Generated clients / versioned contracts A written agreement about how the parts of your system talk to each other, generated from one source. When someone changes something incompatible, it fails on their screen during development rather than for a customer. Queues Async Job Processing Python, FastAPI, Celery, RabbitMQ, Redis △ Idempotent retries / DLQ triage Slow work — reports, imports, emails, payment callbacks — moved out of the customer’s way so pages stay fast. Anything that fails is retried safely and, if it keeps failing, lands somewhere a person actually looks. Database PostgreSQL Under Load PostgreSQL, PgBouncer, Patroni, Prometheus, Linux △ Read scale-out / partitioned writes Making the database keep up as the data grows, usually by finding out what it is actually doing rather than by buying a bigger machine. Cheaper than hardware, and it lasts longer. Caching Multi-Level Caching Redis, NGINX, CDN, Node.js, PostgreSQL △ 4 tiers / explicit invalidation Keeping ready-made answers close to the visitor so pages load fast and the servers do less work. Each layer has an explicit rule for how stale it may be, so nobody is ever shown an old price. Search Search and Document Store Elasticsearch, MongoDB, Python, Kafka, Redis △ Live reindex / change-stream fed Search that finds things by meaning rather than exact wording, and stays current as the catalogue changes. Rebuilding the index happens while the site is open, not overnight with the service down. Web SPA with Server-Side Rendering React, TypeScript, Vite, Node.js, NGINX △ SSR first paint / typed end to end Pages arrive already drawn, so visitors see content immediately and search engines can read it — then behave like an app once loaded. Fixes the common problem of a site that ranks badly because it is empty until scripts run. Design System Component Library Vue.js, TypeScript, Webpack, Storybook, CSS △ 40+ components / WCAG AA One set of buttons, forms and layouts that every page is built from. New screens get assembled instead of drawn from scratch, they look consistent by default, and people using screen readers are not an afterthought. Monitoring Observability Stack Prometheus, Grafana, Loki, Elasticsearch, Alertmanager △ Correlated signals / SLO-based alerts Knowing what the system is doing, and being able to follow one customer’s request all the way through when something goes wrong. The difference between "the site feels slow" and knowing which step is slow, and why. Process Incident Response Grafana, Alertmanager, Ansible, Git, Linux △ Owned rotations / current runbooks A clear answer to who is woken up when something breaks at night, written instructions that match reality, and a review afterwards that produces a fix rather than a culprit.

Professional Milestones

Six shifts the industry has since packaged and sells ready-made. I worked through them before anything was ready-made.