Browse categories
Explore
Fiverr Pro
English
$
USD
I will setup hermes agent webui bug fixes obsidian second brain mcp server claude code
Kuba Kuch
About this gig
Want a powerful self-hosted AI agent without the technical headache? I install and configure OpenClaw and Hermes Agent so you get an autonomous, 24/7 AI assistant running on your own machine or VPS private, secure, and always on.
Need more? I build the complete stack Obsidian second brain,n8n automation,MCP server setup, Ollama, Paperclip AI memory, and multi-agent routing with Claude Code all self-hosted and hardened for security.
What I deliver:
- Hermes Agent deployed on your VPS with multi-agent routing configured
- n8n workflows connected to Hermes for automated task execution
- CRM integration (GoHighLevel, HubSpot, or your platform) leads and updates sync automatically
- Obsidian vault wired in as persistent memory via MCP
- Ollama local model setup for zero-cost inference where it makes sense
- Docker deployment with auto-restart for 24/7 uptime
- MCP server configuration connecting Claude AI to your whole stack
- Set up your Hermes Agent MCP catalog install n8n, GitHub, Linear, Supabase, Notion, HubSpot
- Set up your self-hosted local LLM with Ollama, OpenWebUI or AnythingLLM
ORDER NOW !!!
Get to know Kuba Kuch
Kuba Kuch
AI Automation for Businesses
- FromPoland
- Member sinceJun 2026
- Avg. response time3 hours
Languages
Polish, English, German, Russian
AI Receptionist and Appointment Booking solutions for local businesses that want to capture more leads, answer calls 24/7, and automate scheduling. I build AI voice agents using Vapi, Retell AI, Bland AI, OpenAI, ElevenLabs, Twilio, GoHighLevel, n8n, Make, and CRM integrations to streamline customer interactions. Message me with your business workflow or automation goal, and I'll tell you exactly what I can build for you.
My Portfolio
FAQ
Will this actually reduce my monthly api costs or just move the problem?
Yes it genuinely reduces cost. I route cheap tasks like heartbeats and simple lookups through ollama running locally for free, and use deepseek v3 for real reasoning at a fraction of claude or gpt pricing, reserving premium models only for what actually needs them.
Do I need my own hetzner account or do you provide the server?
Yes you need your own hetzner, contabo, or similar vps account since you own and control the infrastructure directly, and I handle the full setup, hardening, and deployment on it.
Can hermes agent and openclaw run entirely on local models without any cloud api?
Yes this is possible for many tasks. I configure ollama with qwen, llama, or mistral for fully local operation where quality allows, and only call deepseek or claude for tasks that genuinely need stronger reasoning.
Will this fix the token waste and runaway loop problems from my current setup?
Yes. I configure daily spend caps, loopback only gateway access, and proper session management from the start, so you avoid the heartbeat waste and runaway bills that unmanaged setups are known for.
Can this connect to n8n and my existing whatsapp or telegram bots?
Yes. I wire your agent to n8n for scheduled automation and webhooks, and connect it to whatsapp, telegram, discord, or your existing crm so it operates where your business already runs.
Do I keep full ownership and access to everything after delivery?
Yes. Everything runs on your own vps under your own accounts, so you are never locked into a platform or dependent on me to keep it running.
Do you provide a walkthrough so I can manage this myself?
Yes, every delivery includes a full walkthrough covering how the model routing works, how to monitor costs, and how to add new tasks or channels yourself.
