Cost & reliability
Find the line items quietly draining your budget and cut them — right-sizing, autoscaling, bin-packing, killing idle spend — then make deploys boring enough that nobody gets paged at 3am. Boring beats clever.
Kubernetes & Cloud Cost EngineeringCut your cloud bill 30–50%.
Cut your cloud bill 30–50%.
Without touching reliability.
For scaling teams whose AWS or Kubernetes bill is growing faster than revenue. I find the waste, cut it, and keep deploys boring — and I build the systems most engineers only host. Fixed scope, no bodyshopping.
Book a free discovery call30 minutes · no obligation
- 30–50%typical cloud & Kubernetes cost cut
- 6+ yrsAWS, GCP & Kubernetes in production
- Fixedscope & price — approved before I start
- Cost optimization
- Kubernetes
- Cloud migration
- CI/CD & IaC
What I do
Consulting that pays for itself
Kubernetes
Kubernetes — but only if you actually need it. Set up to grow when traffic spikes and shrink (with the bill) when it doesn't, minus the thousands of lines of YAML nobody understands.
Cloud migration
Get onto AWS or GCP without a scary big-bang weekend. We move what's worth moving, skip what isn't, and keep everything running while we do it.
CI/CD & IaC
Make shipping a non-event: you push, it deploys, everyone moves on. The whole setup lives in Terraform — reviewed, versioned, and rebuildable from scratch.
Where this leads: infrastructure you can stop thinking about — so your team gets back to building the product, not babysitting servers.
Beyond ops, I also review architecture, audit AI-built apps, and take ideas to production.See all services →
How it works
From first call to production
- 01
Audit
A free call and a look at your stack. I find the quick wins and the real risks — no jargon, no upsell.
- 02
Plan
A clear, prioritized roadmap with effort, cost, and expected impact. You decide what gets done and when.
- 03
Ship
I implement alongside your team, document everything, and leave you able to run it without me.
Proof I build hard systems
I built a Signal-grade encrypted messenger from scratch
Chatter is a privacy-first encrypted messenger I build solo — one Rust client (Slint) running on Linux, Windows, and Android today, with iOS and macOS planned, on a Go backend with full Signal Protocol end-to-end encryption. When I say I understand your infrastructure deeply, this is the receipt: cryptography to deployment, one person, real correctness stakes. Most DevOps consultants can only host systems like this. I build them.
1 codebaseGo backendRust clientSignal E2EIn development
Who you'd work with
Hi, I'm Alexandru
I'm a DevOps engineer with 6+ years shipping cloud infrastructure on AWS, Kubernetes, and Terraform — and my specialty is cutting cloud and Kubernetes bills without sacrificing reliability. On my team I own security reviews and performance tuning — the same eye for risk, waste, and slow paths I bring to your stack. What sets me apart: I don't just operate systems, I build them from scratch — like Chatter, an encrypted messenger with Signal-grade cryptography I develop solo. That depth is why I find waste and risk other people miss. I care about boring, reliable systems and explain the trade-offs in plain language, not buzzwords.
Questions
Before you book a call
What does a DevOps consultant actually do for my team?
I find what's slow, expensive, or fragile in your infrastructure and fix it — then set things up so your own team can run it after I leave. The goal is to make myself unnecessary, not to lock you into me.
Will migrating to the cloud cause downtime?
No scary big-bang weekend. We move one piece at a time, keep production running the whole way, and roll back the moment anything looks off. Most migrations finish without users noticing.
Do I actually need Kubernetes?
Honestly, often not. If a handful of containers or managed services do the job, that's what you get. Kubernetes only goes in when your scale or complexity actually earns the extra moving parts.
How do you charge?
It starts with a free call and a look at your stack. From there you get a fixed-scope proposal or a day rate — your choice. You approve the plan and the cost before any work begins.
How soon can we start, and do you work with my existing team?
Usually within a week or two. I work alongside your engineers rather than over them, document everything as I go, and hand it over so nothing depends on me being around.
Ready to cut your cloud bill?
Book a free 30-minute call. We'll find the quickest wins in your infrastructure — no commitment.
Book a free discovery callLatest Posts
View all posts →Two Cryptos, One Binary: How Rust Links With C (and How I Got Heap Corruption)
A Windows build of my chat app crashed with no Rust panic—just an access violation deep in C. Tracking it down meant learning how compilation, symbols, and linking actually work underneath cargo.
Two Cryptos, One Binary: How Rust Links With C (and How I Got Heap Corruption) — Read more →The Cat-and-Mouse Game: How Developer Recruitment Scams Mutated in 2026
The 'just read the code' advice is dead. From weaponized VS Code tasks to clipboard-hijacking CAPTCHAs, here is how state-sponsored actors like Lazarus have evolved their tactics to compromise even the most cautious developers.
The Cat-and-Mouse Game: How Developer Recruitment Scams Mutated in 2026 — Read more →From React to Rust: Why I Switched to Native Mid-Project
Building a chat app taught me that every abstraction layer has a cost—and some costs only become clear when you're deep in development.
From React to Rust: Why I Switched to Native Mid-Project — Read more →