I need your help building a workflow in JSON format to use on the n8n platform.
Here’s the project idea:
I will upload videos to Google Drive.
Each time, I want n8n to automatically pick up:
The video
Hashtags
Tags
Thumbnail image
Requirements:
Create an automation that retrieves these files from Google Drive and processes them in the best suitable way.
Very important: I do NOT want the video to be downloaded to the VPS at all, because the videos are in 4K and can reach up to 8GB in size, while my VPS has limited resources (2GB RAM).
So the workflow should rely on streaming or direct processing without full download.
After the process is completed:
Automatically delete:
The video
Hashtags
Tags
Thumbnail image
from Google Drive.
Let me know if you need more details about the data structure or organization.
Apart from that I’m also a full-stack developer with the right Gen AI experience, which makes me a solid plus for your team[but right now only vibecoding]
Check my recent gen ai projects… I built a native Android automation agent too. It’s worth a look:
I can build complex AI automations directly in code, not just inside n8n
I recently started posting my n8n work on YouTube with explanations:
Your biggest challenge isn’t just automation it’s designing a workflow that can reliably process massive 4K video files (up to 8GB) on a lightweight 2GB VPS without downloading them locally or overwhelming server resources.
That’s exactly the kind of systems architecture I specialize in.
My Approach: Zero-Download / Direct-Stream Workflow Architecture
Rather than pulling large video files into n8n memory (which risks RAM overload and VPS crashes), I’d design this workflow as a direct-routing automation system that uses Google Drive API references and streamable links instead of local file handling.
Workflow Structure
1. Google Drive Intake Monitoring
n8n watches a designated Google Drive folder for each new upload package
Detects:
Video file
Thumbnail
Hashtags file
Tags file
2. Metadata Extraction (Lightweight)
Tags/hashtags are parsed directly from text, JSON, or structured file formats
Minimal server load since only lightweight metadata is processed
3. Zero-Footprint Video Handling
The video is never downloaded to your VPS
n8n generates a secure Google Drive direct-access or stream link using API methods
That link is passed directly to the target platform/system for processing
Result:
Your VPS acts purely as an orchestration layer — not a storage or processing machine.
4. Success Validation Layer
Workflow confirms successful transfer/processing before cleanup
Prevents accidental deletion if the destination process fails
5. Automated Cleanup Once confirmed:
Delete video
Delete thumbnail
Delete hashtags
Delete tags
from Google Drive automatically
Why This Architecture Works
Benefits:
No RAM overload
No 8GB local storage usage
Faster execution
VPS-safe
Scalable for repeated uploads
Deliverables
I can provide:
Clean, production-ready n8n JSON workflow
Google Drive API integration
Error handling + validation logic
Organized, documented structure for easy deployment
I focus on building workflows that operate efficiently under real infrastructure limits — not just basic automations.
If you’d like, I can also help optimize this further depending on your final destination (YouTube, social scheduler, cloud editor, storage transfer, etc.), so the routing method matches the platform perfectly.
Best regards, Agbaje Olarewaju
AI Automation Architect
System Design • Workflow Automation • Scalable Operations
I came across your n8n workflow requirement and would love to help.
I’m Pranay Pise, with 3+ years of experience building advanced n8n automations, API integrations, and scalable workflow systems.
Your use case can be handled efficiently using Google Drive triggers, metadata processing, stream-based file handling (without full VPS download), and automated cleanup workflows — ideal for large 4K files with limited server resources.
The no-download constraint is the right call given your VPS specs — totally doable but the architecture depends on one thing you haven’t mentioned: what happens to the video after retrieval? Are you uploading it to YouTube, sending it somewhere else, transcribing it?
The reason it matters: for something like YouTube, I can pass the Google Drive share URL directly to the YouTube API without ever pulling the file to your VPS. For other destinations the approach changes. The hashtags, tags, and thumbnail I can handle cleanly regardless — those are small files, no memory issue.
Once you confirm the destination I’ll send you the full workflow JSON. Shouldn’t take long to build once that’s clear.
I’ve built almost the same flow recently — n8n on a low-resource VPS handling large media via Google Drive direct streaming, no local download, full cleanup after processing.
Here’s the approach: • Google Drive trigger watches the upload folder for new video + asset bundle • Files matched by naming convention or folder structure (video + thumbnail + hashtags.txt + tags.txt) • Direct Drive file streaming via download URL passed to the destination platform — n8n only handles metadata, not the binary • Hashtags + tags + thumbnail processed normally (small files, no VPS strain) • Post-success cleanup: delete all 4 assets from Drive via Drive API, with retry + error logging
One quick question before I scope: where is the video being sent after pickup? YouTube, social platforms, custom API, etc. — that determines whether streaming pass-through is possible (YouTube/most APIs support it) or whether we need a different upload pattern. Affects the build.
Let’s schedule a call to discuss the details further.
Moving 8GB 4K videos on a 2GB RAM VPS is a specific challenge because n8n’s standard nodes will crash your server by trying to load the file into memory. I can solve this by building a streaming pipeline that bypasses your RAM entirely.
How I’ll handle your project:
Direct Streaming: I will use a custom Node.js script (JavaScript) inside n8n to “pipe” the data directly from Google Drive to your destination. This ensures the video never actually “sits” on your VPS, keeping your 2GB RAM safe.
Asset Logic: I’ll build the workflow to identify and group the video, thumbnail, and text files (tags/hashtags) so they are processed as one complete package.
Safe Cleanup: I will implement a check-step so that the Google Drive Delete node only triggers after the destination server confirms the upload is 100% successful.
Architecture: As an automation specialist, I focus on building “low-resource” systems that are stable and don’t require manual restarts.
Why me?
I’m a developer who knows the “under the hood” mechanics of n8n and APIs. I don’t just drag nodes; I write the code needed to handle heavy data transfers.