DeepSeek V4 Pro + Hermes Agent: Build n8n Workflows via Chat

#n8n #HermesAgent #DeepSeekV4 #AIAgent #AIAutomation #Supabase #TelegramBot #SlackBot #n8nTutorial #NoCode
→ Supabase: https://supabase.plug.dev/fs3vilS
→ Hermes Railway template: railway.com/deploy/hermes-agent-by-no-1
→ Supabase agent skill: supabase.com/blog/supabase-agent-skills
→ Perplexity API: perplexity.ai
→ n8n: https://n8n.io
→ Previous video (Hermes self-improving mechanic): https://youtu.be/W_ZgH0WPayo

📧 FREE Weekly AI Automation Newsletter:
Get weekly tutorials, free workflow templates, and the latest AI news straight to your inbox
👉 https://aifornoncoders.com/newsletter

Build a Telegram and Slack customer support bot using Hermes Agent and DeepSeek V4 Pro — without touching the n8n canvas. Every workflow is built through conversation.

Hermes connects to n8n's MCP server and builds workflows from a single prompt. DeepSeek V4 Pro powers it: 1.05M token context, Opus 4.6 performance at a fraction of the cost — near-zero when the cache hits, which it does often. The more workflows Hermes builds, the sharper it gets.

What you'll build:
→ Customer support bot on Telegram + Slack
→ Live order and product data from Supabase (built by asking — no SQL)
→ Perplexity AI search fallback for anything the database doesn't have
→ Full AI agent with window memory across conversations
→ Everything wired together through Hermes chat — no n8n canvas work

──────────────────────────────────────
CHAPTERS
──────────────────────────────────────
0:00 Demo — the complete system running
0:30 What you're building + Hermes overview
1:00 Deploy Hermes via Railway template
3:00 Connect n8n MCP server
5:00 Install Supabase skill + build database by asking
9:00 Build the Telegram workflow with one prompt
11:00 Wire Supabase into the AI agent
13:00 Add Perplexity AI search
15:00 Add Slack with OAuth2
17:00 Full system test + real cost breakdown
18:30 Honest friction — what breaks and what to plan for

──────────────────────────────────────
TOOLS USED
──────────────────────────────────────
• DeepSeek V4 Pro — $0.435/M tokens standard | $0.003625/M cache reads
• Hermes Agent — 133K GitHub stars, fastest-growing AI agent repo 2026
• n8n (self-hosted or cloud)
• Supabase
• Perplexity AI
• Telegram Bot API
• Slack OAuth2
🤝 Join the AI Automation Engineering Community:
Go deeper — interact with me and the community, get feedback on your builds, and stay ahead of the curve
👉 https://www.skool.com/ai-automation-engineering-3014

Видео DeepSeek V4 Pro + Hermes Agent: Build n8n Workflows via Chat канала Derek Cheung | AI Agents Automation