Skip to content

n8n Workflow

This page collects notes about my n8n workflows and how Iā€™m using them in small automation projects.

šŸš€ Quick Docker setup for n8n

Follow this graphical installation flow to set up n8n quickly using Docker:

Step 1: After Docker installation, go to "images" search for n8n, and pull the image. Step 1

Step 2: Check the pulled image. Step 2

Step 3: Go to "Volumes" and "Create a Volume". Step 3

Step 4: Check the created volume. Step 4

Step 5: Run the new container with the n8n loaded inside by clicking the "Run" button. Step 5

Step 6: Now you are inside n8n and click the "Create Workflow" to start. Step 6

Step 7: Here is a basic structure of n8n workflow. From here you can start your workflow creation. Step 7

After finishing these steps, open http://localhost:5678 in your browser and complete the initial setup wizard to start building your workflows.

šŸ“ø Project 1 ā€“ AI Background Replacement and Selfie Booth

An AI selfie booth that takes photos of users and generates any background, or transforms the user into any famous character. The image shows a passerby taking a photo in front of an open-style selfie booth. The front screen (acting like a mirror) displays a real-time image of the passerby transformed into Captain America.

AI Avatar / AI selfie booth

šŸ“ Project 2 ā€“ Student Assignment Grading & Exercise Recommendation System

The second n8n project is a student assignment grading system:

Goal - Students take photos of their assignments (paper homework, exercises) and upload them. - The system automatically recognizes the questions and answers, and provides grading feedback. - Based on the types of errors, it recommends a few similar exercises to help reinforce weak areas.

n8n Workflow Idea - Trigger: Students upload homework photos via a form / Webhook. - Processing pipeline: 1. OCR / Vision model recognizes the questions and the student's answers. 2. Call a large language model (e.g., GPT family) to grade and analyze against the standard answers. 3. Based on the knowledge gaps, generate 3ā€“5 similar exercises from a question bank or the LLM. 4. Combine the grading feedback + explanations + new exercises into a report, returning it via email or a frontend page.

What kind of product should this be? - Standalone: A small tool for personal use by teachers (photo ā†’ grading ā†’ generate PDF of exercises). - Web version: Students self-upload and get real-time "grading + exercises" feedback. - Integrated with existing LMS / homework systems: Connect via n8n's HTTP / API nodes into existing systems.

I will continue to add specific n8n node configuration screenshots and exported JSON files here later, so you can easily import and reuse the workflows. End Patch}...