Train and run a deep neural network via webhooks using pure code nodes

by Alex Raj ✓ · 3 views · created 2026-08-25 · Engineering, AI Summarization

What this workflow does

This workflow has 45 nodes and runs when an HTTP POST request hits its webhook URL (plus 1 other trigger). It uses only n8n's built-in nodes.

How it runs

Trigger: When an HTTP POST request hits its webhook URL (plus 1 other trigger).

1. Run Backpropagation (Code): runs custom JavaScript code
2. Initialize Layer 1 Node (Code): runs custom JavaScript code
3. Initialize Layer 1 Node (Code): runs custom JavaScript code
4. Initialize Layer 1 Node (Code): runs custom JavaScript code
5. Initialize Layer 1 Node (Code): runs custom JavaScript code
6. Initialize Layer 1 Node (Code): runs custom JavaScript code
7. Initialize Layer 1 Node (Code): runs custom JavaScript code
8. Initialize Layer 1 Node (Code): runs custom JavaScript code
9. Initialize Layer 1 Node (Code): runs custom JavaScript code
10. Initialize Layer 1 Node (Code): runs custom JavaScript code

…plus 33 more steps. See the full node list below.

Before you upgrade to n8n 3.0

Nodes used (4)

• Coden8n-nodes-base.code
• Mergen8n-nodes-base.merge
• Webhookn8n-nodes-base.webhook
• Respond to Webhookn8n-nodes-base.respondToWebhook

From the author

Quick overview This workflow exposes two webhooks to train and run inference for a 3-hidden-layer neural network built entirely with n8n Code and Merge nodes, persisting learned weights in workflow static data and returning class probabilities, predictions,…

View the original on n8n.io → Check your own copy