Build new projects
From initial idea and architecture to a working, deployable system with infrastructure, observability and documentation.
Senior DevOps & AI Automation Engineer
I build and modernize software systems with AI-assisted development.
I combine long-term DevOps and infrastructure experience with AI-assisted engineering to create new projects, take over existing systems, automate operations, and turn ideas into working environments.
About
My background is in Linux systems, infrastructure automation, networking, virtualization, containers, CI/CD and platform operations. Over time I expanded this work into complete AI-assisted project development: defining architecture, breaking work into RFC-sized tasks, using AI coding agents for implementation, validating results, and taking responsibility for integration and operations.
What I do
Take over existing systems
Understand unfamiliar code and infrastructure, identify operational risks, stabilize the environment and establish a maintainable workflow.
AI-assisted engineering
Use ChatGPT and Codex as engineering tools for architecture, RFC preparation, implementation, refactoring and iterative delivery.
DevOps & automation
Linux, Docker, Terraform, Ansible, GitLab CI/CD, monitoring, networking, virtualization and reproducible environments.
Selected projects
JobRunner
A web application for controlled execution of Docker-based jobs with live logs, PostgreSQL, lifecycle management, Prometheus, Loki and Grafana.
Scrapper
A Dockerized public-data collection project for Google Maps search results, later integrated with JobRunner and observability workflows.
Codex1
My first Codex project. It started as a learning experiment and evolved into Terraform + Ansible automation for Hyper-V environments on Windows 10.
gpt_bot
An early experiment in connecting the OpenAI API, Git and automated file modification — a precursor to my later AI-assisted development workflow.
AI Journey
ChatGPT
Manual collaboration
Prompts, generated files and iterative technical work.
Git
Safety and reproducibility
Version control became essential after repeated AI-generated changes occasionally broke working code.
gpt_bot
My own automation layer
I experimented with OpenAI API, prompt history and automatic file updates committed to Git.
Codex
Agent working directly with repositories
The workflow shifted from receiving code to delegating bounded implementation tasks.
RFC workflow
Architecture → task → implementation → verification
Small RFCs, Git history and regression checks made AI-assisted development more predictable.
Core technologies
LinuxDockerDocker ComposeTerraformAnsible
GitLabCI/CDPythonBashPowerShell
NginxPrometheusLokiGrafanaProxmox
Hyper-VWireGuardNetworkingChatGPTCodex