Case Study

One video in,
three platforms out.

Automated content distribution system that converts YouTube music videos into 60-second vertical clips and posts them across TikTok, Instagram Reels, and Pinterest via Buffer API with platform-specific captions and hashtags.

Status Live · Internal
Client JQ Internal
Category Content Distribution
Built 2026
At a glance

What is Tidesstunes Social Pipeline?

The Tidesstunes Social Pipeline is an automated content distribution system built for the Tidesstunes music project (relaxing piano and ocean music). It downloads YouTube videos, encodes the best 60-second segment into vertical 9:16 clips, uploads to Cloudinary CDN, then distributes across TikTok, Instagram Reels, and Pinterest via Buffer's API with platform-specific captions, hashtags, and formatting. 17 videos processed, 39+ posts queued across 3 platforms.

The Problem

What was broken.

Tidesstunes publishes relaxing piano and ocean music on YouTube. Every new video needs to reach TikTok, Instagram Reels, and Pinterest to grow the audience. Each platform has different requirements: vertical 9:16 format, different caption lengths, different hashtag strategies, different posting conventions.

The manual version of this workflow takes 30+ minutes per video. Download from YouTube, open a video editor, find the best 60-second segment, export at 9:16, then log into each platform separately, write platform-specific captions, add the right hashtags, post. Multiply by 17 videos and that is an entire day of repetitive work that adds no creative value.

Scheduling tools solve the posting part, but not the conversion part. They do not download from YouTube, do not encode to vertical format, do not select the best segment, and do not generate platform-aware captions. The gap is the full pipeline from source video to posted content.

The Approach

What was built.

The system is a three-stage pipeline. Stage one handles source acquisition and encoding: yt-dlp downloads the YouTube video, FFmpeg selects the best 60-second segment (not the first 60 seconds, but the musically strongest section) and re-encodes it to 9:16 vertical format optimised for mobile viewing.

Stage two handles cloud hosting: each processed clip uploads to Cloudinary CDN, which provides the public URL that Buffer needs for posting. Cloudinary handles format optimisation and delivery.

Stage three is distribution: the Buffer API receives each clip with platform-specific metadata. TikTok gets short, hashtag-heavy captions. Instagram Reels get slightly longer captions with engagement hooks. Pinterest gets keyword-rich descriptions optimised for search discovery. Rate limiting and retry logic prevent API throttling across batch runs.

How It Works

Architecture in plain English.

01
YouTube download
yt-dlp pulls the source video from YouTube at the best available quality. Metadata (title, description, duration) is extracted and stored for caption generation.
02
Segment selection
The system identifies the strongest 60-second segment based on audio energy analysis, not just the first minute. This ensures the clip represents the best part of each track.
03
Vertical encoding
FFmpeg re-encodes the selected segment to 9:16 vertical format (1080x1920) with centre-crop framing. Output is optimised for mobile playback: H.264, AAC audio, fast-start flag for instant play.
04
Cloudinary upload
Each processed clip uploads to Cloudinary CDN via their Python SDK. The returned public URL becomes the media source for all three platform posts.
05
Platform-specific captions
Each platform gets tailored copy. TikTok: concise with trending hashtags. Instagram: engagement-focused with a call to action. Pinterest: keyword-rich for search discovery, assigned to the "Ocean Piano Meditation Music" board.
06
Buffer API distribution
Buffer's API receives three separate posts per video (one per platform) with the correct caption, hashtags, and media URL. Rate limiting spaces posts to avoid spam detection. Retry logic handles transient failures.
Try It

See it in action.

Interactive walkthrough

Walk through the full pipeline: selecting YouTube videos, analysing audio segments, encoding vertical clips, uploading to Cloudinary, generating platform-specific captions, and queuing posts across TikTok, Instagram Reels, and Pinterest.

Tidesstunes Social Pipeline dashboard showing source video library with real YouTube thumbnails, 8-step pipeline timeline, and channel branding
Launch Interactive Demo
Pipeline flow

Source video enters, three platform-ready posts come out. Every step is automated.

YouTube
Encode 9:16
Cloudinary CDN
Buffer API
TikTok
Instagram
Pinterest
Platform distribution

Each platform receives tailored content. Same video, three different posts.

Stack

Built with.

Python 3 yt-dlp FFmpeg Cloudinary SDK Buffer API TikTok Instagram Reels Pinterest
Outcomes

What changed.

17 videos processed
39+ posts queued
3 platforms served
~2 min per video (vs 30+ manual)

The biggest win is not speed but consistency. Every video now reaches all three platforms with properly formatted content instead of being posted to one and forgotten. The platform-specific caption generation means each post reads native to the platform it appears on, not like a cross-posted afterthought.

Free Consultation

Want a system
like this one?

Book a free 30-minute call. We map your situation, scope a similar build, and agree on a fixed quote before anything starts.

Book Free 30-min Call