My Journey with n8n Workflow Has Started

Adapted from my LinkedIn article: My journey with n8n workflow has started..

I experimented with many AI tools, but most did not stick in day-to-day use. n8n clicked because it solves a practical problem: event-driven automation with low friction.

Why n8n #

n8n is not just “an AI tool.” It is a workflow automation engine.

In simple terms:

• an event arrives (message, webhook, file, API)
• a workflow runs
• output is generated and can trigger another workflow

My first use case was a WhatsApp assistant backed by local Ollama models.

Lab setup #

• one spare PC running n8n
• one laptop with GPU for Ollama
• Docker-based deployment for n8n

Ollama installation was straightforward. n8n setup via Docker quickstart was also simple.

Real challenge: public webhook endpoint #

WhatsApp webhooks require a reachable HTTPS endpoint. My home setup had no static public IPv4.

I evaluated:

1. Cloudflare Tunnel
2. ngrok
3. DuckDNS + Let’s Encrypt

I chose the DNS + certificate route because it was scriptable and repeatable for future instances.

Integration notes #

Meta setup was the trickiest part: receive flow and send flow use different IDs/tokens. After baseline echo tests, I connected n8n AI node + Ollama and added memory support.

One useful expression pattern in n8n:

{{ $json.messages[0].text.body }}

This avoided hardcoding and passed message content across nodes dynamically.

Next steps I planned #

• scale the setup to cloud
• automate replication across n8n instances
• selectively move credentials while exporting/importing workflows

Takeaway #

n8n gave me a practical bridge between local AI models and real messaging workflows. It turned experimentation into something operational.