01
System audit
To find bottlenecks and see where stability is being lost.
// Stable infrastructure instead of firefighting
Is your infra breaking down and held together with workarounds?
I remove outages and manual routine from infrastructure so you stop losing money on incidents, can grow the product calmly, and do not depend on manual actions or human error.
Based on work across dozens of production projects: I measure outcomes by three metrics — incident frequency, mean time to recovery, and share of manual operations. I usually handle 2-3 projects at a time. More about BigMike.help →
22+ years of experience
140 reviews · 5.0 out of 5
Clear plan before work starts
I do not break what already works
DevOps + Python automation
BigMike.help is a DevOps and Python automation service for production systems. I help startups, product teams, and small businesses stabilize infrastructure, improve CI/CD, monitoring, backups, and remove brittle manual operations.
// Case studies · 74 projects solved
Systems already brought back under control
Short and concrete: problem, solution, and end state.
Case
Server breached via PostgreSQL
❌ Problem
PostgreSQL was exposed to the internet with the password postgres and no firewall: the first attack came within 6 hours, the breach in under a day
✅ Solution
traced the attack chain: a Monero miner, a C2 agent, and event triggers that kept recreating a superuser role
📈 Result
it is clear how they got in and stayed; three rules plus connection and DDL logging close off this class of attack
Case
Managing 366 servers
❌ Problem
every new server took 1–1.5 hours to provision and depended on memory and manual steps
✅ Solution
built Ansible roles, Semaphore, and inventory to manage 366+ servers across seven countries
📈 Result
a new node is live in 8–12 minutes without manual SSH, and the infrastructure is fully reproducible
Case
Log diagnosis in Telegram
❌ Problem
there was no monitoring: on failure, logs were pasted into ChatGPT by hand, after the client had already called
✅ Solution
on a Grafana alert, n8n pulls 15 minutes of pre-incident logs from Loki, an LLM analyzes them, and the answer goes to Telegram
📈 Result
the outage alert arrives with a root-cause analysis attached, without Zabbix, Prometheus, or in-house DevOps
// Reviews · 140 ratings, 5.0 out of 5 average
The result: stable operations within the first days
Clients come in with outages, manual routine, and no visibility. They leave with infrastructure that behaves predictably.
★ 5/5
«I needed to get n8n, Redis, and the database working. I had hired another contractor before and everything kept breaking. I hired Mikhail, and the next day everything was working quickly, like clockwork!»
★ 5/5
«Thanks to Mikhail for his responsiveness. We spoke by phone; he explained how I could do it myself. This is my second time reaching out — everything’s great and prompt.»
★ 5/5
«The project was large and complex. Everything was done exactly as I had imagined, down to the smallest details. Mikhail paid close attention to all requirements and functionality. The project is well structured and, …»
// Where to start
We can start with an infrastructure review
02
Consultation
To quickly understand what to do next and what should not be touched.
03
Improvement plan
To get a clear implementation sequence without chaos.
Free · no sign-up
Run a free siteDoc check
Not ready to write? Check your site yourself
siteDoc is a free online diagnostic: it checks your website's configuration, speed, and security and shows the weak spots.
// How the work goes
How the work goes
You describe the task and the current problems.
I review the infrastructure and find the weak points.
I give you a concrete work plan with priorities.
I implement the fixes and carry the system through to a stable state.
// CTA
Want to stop losing time and money on problematic infrastructure?
Send the task in Telegram or via the form. I will review the project, show the weak points, and tell you what to fix first — to stop losing money and remove recurring incidents.
- stable infrastructure without constant failures
- predictable releases without stress
- minimal manual work and less routine
- a system that can scale calmly
Write in Telegram
I usually handle 2-3 projects at a time. I take new ones as capacity opens up