Browse categories
Explore
Fiverr Pro
English
$
USD
I will deploy self hosted ai stack openwebui openhands openclaw litellm and others
Said E
Level 2
About this gig
I will deploy a self-hosted AI stack that connects to your paid LLM APIs
Need AI infrastructure that works with the models you already pay for? I deploy a self-hosted stack on your VPS a unified gateway, chat UI, agents, and knowledge tools all wired to OpenAI, Anthropic, Google, or any provider you use. No lock-in, no per-seat SaaS fees. Local models optional as fallback.
Core stack: LiteLLM (unified gateway, usage/cost tracking), Open WebUI (chat), Qdrant (vector DB/RAG), SearXNG (private search), Nginx Proxy Manager, Portainer, Vaultwarden.
Basic $90 Core stack, custom domain, SSL, tested.
Standard $150 Adds n8n, OpenHands (self-hosted coding agent), Flowise (visual agent builder), Uptime Kuma, automated backups.
Premium $200 Adds Open Notebook (self-hosted research assistant), OpenClaw (agent runtime for Slack/WhatsApp/Telegram automation), Keycloak SSO, one login for everything.
Every package includes:
Docker Compose, clean architecture
Custom domain + SSL
Full testing before delivery
Setup docs
14-day post-delivery support
Requirements: VDS 24GB+ RAM, AlmaLinux 9, SSH access, subdomains mapped to server IP. Bring your own API keys I don't resell LLM access.
Get to know Said E
Said E
Linux Server and Automation Expert
Level 2
- FromMorocco
- Member sinceOct 2014
- Avg. response time1 hour
- Last delivery3 weeks
Languages
English, Arabic, French
With 10+ years in Linux/DevOps, I build self-hosted AI infrastructure for businesses that want it done right — not another weekend of free tutorials and broken containers.
I deploy: LiteLLM, Open WebUI, Qdrant, Flowise — connected to your paid LLM providers (OpenAI, Anthropic, Google). OpenHands for self-hosted coding agents. OpenClaw for agent automation across Slack/WhatsApp/Telegram. n8n with custom dashboards. Keycloak SSO and server hardening.
Everything runs on YOUR server — zero lock-in, no per-seat fees, tested and documented.
Platforms: Contabo, Hetzner, DigitalOcean, Vultr, AWS.
My Portfolio
Other AI Development Services I Offer
FAQ
Do I need my own LLM API keys?
Yes. You bring keys for whichever provider you use (OpenAI, Anthropic, Google, etc.). I don't resell or mark up API access — you pay the provider directly, and LiteLLM tracks your usage/cost in one place.
Can I switch or use multiple LLM providers later?
Yes. LiteLLM is provider-agnostic — you can add, remove, or split traffic across providers without redeploying the stack.
Is my data private if I'm using OpenAI/Anthropic through this?
Your infrastructure, users, chats, and documents stay on your own VPS. Prompts sent to a provider are still subject to that provider's own data policy — I'll point you to the relevant terms for whichever provider you pick.
What if I don't have a VPS yet?
Message me first — I'll help you pick a provider/spec that meets the requirements (24GB+ RAM, AlmaLinux 9) before you buy.
Can I upgrade tiers after delivery?
Yes, the difference between tiers can be added on afterward as a custom order — no need to redo the base stack.
Do you provide ongoing maintenance after the 14 days?
Not included by default, but available separately if you want it — happy to quote once I see your setup.
What if my server doesn't qualify?
No charge for checking — send me the specs first and I'll confirm in minutes.
