DevOps for small IT teams.
Servers, deployments and monitoring
without an in-house engineer
15 years of experience. I take care of your infrastructure while your team builds the product. Hourly or fixed-fee support.
Sound familiar?
If you recognize yourself in at least one of these, let's talk.
Need to move off a provider?
A provider shuts down servers, changes terms or becomes unavailable in your region. You need to migrate without downtime.
Paying for idle cloud
Servers under 20% utilization, zombie environments, forgotten dev stands. Half of the cloud bill goes to waste.
You learn about outages from customers
No monitoring, alerts stay silent, bugs are traced through logs after the fact. Bottlenecks surface too late.
Inherited someone else's infra
The previous DevOps left, there are no docs, Kubernetes is three years out of date, and there is no network map either.
Server and Kubernetes Administration
The six things clients come to me for most often. Pick one or combine them.
System Acceleration
Linux server bottleneck analysis, database and web server optimization
✓ Faster load times = more sales
Cloud Cost Savings
Cost audit, right-sized servers, auto-scaling
✓ Same functionality for less money
24/7 Monitoring
Proactive problem detection, alerting and auto-recovery
✓ 99.9% SLA and peace of mind
DevOps Automation
CI/CD pipelines, Infrastructure as Code, deployment automation
✓ Deployments in 30 minutes instead of 4 hours
Backups
Automated backups, data replication, disaster recovery strategies
✓ No data loss, recovery in minutes
Infrastructure Security
Secure password storage, isolated server access, regular vulnerability scanning
✓ Zero security incidents
How we work together
Five steps from the first message to infrastructure that works.
Introduction
Write to me on Telegram or email. We clarify the task and I answer your questions. A call if needed.
Contract & Access
We sign a contract, NDA if required. You provide access to servers and cloud accounts.
Task Tracker
I work in your tracker or set up mine - Jira, Linear or Notion.
Execution
Every task is agreed and estimated before start. Status and hours spent are visible in the tracker.
Report
Weekly: what was done, hours spent, what's next. Closing documents at the end.
Case Studies
Projects where I was responsible for the infrastructure, and what it gave the business.
Global-Media.info / IPTV video service
Optimization and automation of video streaming servers
- ✓~100 TB of traffic per month on optimized infrastructure
- ✓New stream node in minutes via Ansible
- ✓Monitoring with alerts and daily backups
- ✓Telegram bot for service and billing management
MomsLab.app / App for moms
Infrastructure support and optimization (in progress)
- ✓HA proxy on Caddy with DNS failover via Route53
- ✓Self-hosted SSO on Authelia + LLDAP
- ✓Grafana + Prometheus + Loki with alerts to Telegram
- ✓Backups to S3 with restore testing
CoreMetrics / ML platform
How we migrated ML infrastructure to the cloud and built delivery into the client's internal perimeter
- ✓Migration to Selectel from three home servers to fault-tolerant infrastructure
- ✓CD into the client's internal perimeter via VPN and Git mirroring
- ✓Monitoring and alerts in Pachka
- ✓Backups to S3 with restore testing
BlueWhite / Autonomous tractors
IoT infrastructure for managing a fleet of 100+ autonomous tractors
- ✓Isolated field network: WireGuard + mTLS tractor authentication
- ✓EKS cluster with CI/CD via CodeBuild and ArgoCD
- ✓Real-time video from tractors over GSM
- ✓ML pipeline on AWS SageMaker for perception models
Pricing
Three formats: a one-off audit, hourly work or monthly support.
Audit
Infrastructure map and action plan
one-off · up to a week · depends on infrastructure size
- Map of servers, services and network
- Cloud cost and utilization analysis
- Backup check: do they exist and do they restore
- Vulnerability and access review
- PDF report: what is critical, what can wait, what it costs to fix
Hourly
When you need targeted work without commitment
billed in 30-minute increments · a typical task takes 1-3 hours
- Server and cluster migrations without downtime
- Backups with restore testing
- Incident investigation: logs, metrics, root cause, fix
- Pipeline from commit to production
- Monitoring and alerts to your messenger
Support
When the system already works and you need an engineer on call
/ month · task work billed separately at the hourly rate
- Incident response within 4 business hours
- Regular questions answered the next business day
- I watch monitoring, alerts and backups
- Available on messengers
Technologies
What I use on projects, from clouds to automation languages.
Cloud
Orchestration
CI/CD & IaC
Monitoring
Databases
OS & Network
Security
Storage
Languages
Reviews of a DevOps Engineer
Real reviews from Avito and Profi.ru
"We worked with Ilya as a DevOps engineer on an ML project. During our collaboration he set up the infrastructure and cluster, configured monitoring and CI/CD processes for the team. We went through several large service migrations - he handled them independently. He finds solutions where there are no ready-made answers. Highly recommended."
Daniel
April 28, 2026
"Everything was great - the specialist handled a complex task related to the Cloudflare service being blocked. The deal was successfully completed."
Artem
January 26, 2025
"A very detailed and accurate audit report. Well-structured and easy to understand. A very experienced specialist."
Roman
March 4, 2026
Frequently Asked Questions
Short answers to what people usually ask before we start.
Is this suitable if we only have one server?
Yes. Even with a single server you can reduce cloud costs, set up proper monitoring and automate deployments. Infrastructure size doesn't matter.
How long does an audit take?
Usually up to a week, depending on infrastructure size. You receive a detailed report: what I found, what is critical, what can wait and how much it costs to fix.
Do you work with companies outside Russia?
Yes, I work with international clients. We sign a service agreement, communicate in English and use AWS, GCP, Hetzner or any cloud you prefer.
How quickly can you start?
Usually within 1-3 business days after the task is agreed and the contract is signed.
What if I only need a one-off task, not ongoing support?
One-off tasks are fine. There is no minimum volume - just reach out and we'll discuss.
Can you help migrate between clouds or off AWS / GitLab?
Yes, this is one of my main areas. I help choose alternatives for cloud, Git hosting and monitoring, and migrate without downtime.
We have 20+ servers, 10 developers and a lot of tasks. Will this work?
In that case a larger DevOps team or an in-house engineer is a better fit - that volume is beyond what I offer. Write to me if you're unsure: I'll tell you honestly whether I can help.
We need a custom solution - do you do development?
Yes. I'm a professional developer: I understand networks and servers and write Go, Python, JavaScript, PHP and Bash. I can build a prototype or a small complete product to your requirements, or dig into your existing application code.
Can you design application or infrastructure architecture?
Yes. Based on your requirements I'll prepare an architecture that accounts for load, scalability and budget - as a diagram and documentation.
What happens if you get sick or we need to part ways?
I document all work and manage infrastructure as code - Terraform, Ansible and detailed READMEs. Any DevOps engineer can quickly pick up and continue without losing momentum.
Ilya Manyahin
Saint Petersburg · remote worldwide
Started in 2008 as a sysadmin at Demos, an internet provider. The next 8 years: full-stack development at TopGet, YadWire, Affilomania and Zoominfo. From 2020 to 2023: DevOps Manager at Bluewhite, an Israeli autonomous tractor startup: cloud infrastructure, an IoT fleet of 100+ devices and a team of engineers. Independent DevOps engineer since 2024.
What makes me different
I document everything: any engineer can pick up the infrastructure without losing context
I understand both application code and servers, so I find the root cause of problems faster
International projects in Israel, the USA and Russia: used to different business cultures