Reproducible or it doesn't ship.
Every environment is rebuilt from code and images, so there are no snowflake servers and nothing drifts.
ROLE REMOTE • WORLDWIDE • FULL-TIME OR CONTRACT
DevOps & Cloud Infrastructure Engineer
L5 CLOUD INFRA • L6 CONTAINERS • 1,800+ LIVE SITES
I've designed and run private bare-metal cloud since 2015. Today it hosts 1,800+ live websites. I build deployments that repeat exactly, isolate failures and protect themselves.
DESCEND INTO THE INFRASTRUCTURE LAYERSInfrastructure is a product.
I run it like one.
Most teams rent cloud capacity without understanding how it fails. I design infrastructure from the hardware up, so every layer has a known failure mode, an owner and a recovery path.
- Self-hosted cloud at scale: multi-node Linux VPS topologies, DNS routing clusters and automated TLS.
- Container orchestration: Docker Swarm, immutable images, zero-downtime rolling upgrades.
- Edge and routing: Traefik reverse proxies, zero-trust TLS, centralized DNS.
- Security built into the pipeline: honeypots, IP quarantine and least-privilege admin control.
- Data layer: PostgreSQL, Neo4j and ChromaDB with tested backups.
1,800+ live sites.
Zero hyperscaler markup.
Since 2015 I've built and operated a private bare-metal cloud that hosts a large portfolio of client websites, with no dependency on AWS, GCP or Azure.
Centralized DNS routing clusters, multi-node Linux VPS topologies, automated TLS and Traefik edge proxies keep it up, and the cost stays a fraction of hyperscaler pricing.
Immutable deployments.
Hardened execution kernels.
Infrastructure must be repeatable, sandboxed and able to defend itself. I deploy Docker Swarm container clusters paired with Vivek DevOps, my custom WordPress security layer.
It includes kernel-level honeypot traps, God Admin access control, automated IP quarantine and zero-downtime rolling upgrades, backed by PostgreSQL and ChromaDB.
Stateful services,
self-hosted and secured.
JARVIS runs on my own infrastructure: Neo4j and PostgreSQL behind Traefik, with HTTPS and SSE endpoints serving several AI clients at once.
It shows the full lifecycle I own, from provisioning and TLS to API-key auth, encrypted secrets and upgrades that keep the data intact.
METHODHOW I RUN INFRASTRUCTURE
Operating constraints
Failure is designed for.
Rolling upgrades, health checks and isolation mean one failing service never takes down its neighbours.
Security lives in the platform.
TLS, quarantine and access control are part of the platform from day one, not a ticket for later.
The toolkit.
2015 RUNNING SELF-OWNED CLOUD IN PRODUCTION SINCE
FAQDEVOPS QUESTIONS
Frequently asked
What DevOps experience do you have?
Since 2015 I've designed and operated private bare-metal cloud infrastructure that now hosts 1,800+ live websites. It runs on multi-node Linux VPS topologies, DNS routing clusters, Traefik reverse proxies, automated TLS and Docker Swarm with zero-downtime rolling upgrades.
Can you move a team off AWS, GCP or Azure?
Where the economics and risk justify it, yes. I've run production on self-owned infrastructure for over a decade, and I can design hybrid or fully self-hosted setups, or harden and simplify an existing hyperscaler deployment.
Do you work remotely?
Yes. I'm based in Gurugram, India and work remotely with teams worldwide, full-time or on contract.
How do I hire you?
Email [email protected] or connect on LinkedIn.
ROLESOTHER WAYS I'M HIRED
Other roles
02
AI Engineer & AI Agent Developer
Voice and chat agents, MCP integrations, knowledge-graph memory.
03Full-Stack Developer (React / Next.js)
Fast React, Next.js and Three.js products on Node.js and Python.
04Cybersecurity & SOC Engineer
Distributed threat intelligence, honeypots, adaptive SOC.
05Solutions & System Architect
End-to-end ownership across all ten layers.
INITIATE ENGAGEMENT
Hiring a DevOps engineer?
Open to full-time remote roles and contract infrastructure work, worldwide.