Back to projects
Professional
Video Remotion
Affaq Ahmed · Backend Engineer · DevCache · AI / May 31, 2026
- Node.js
- FFmpeg
- Remotion
- Redis
- BullMQ
Overview
Video Remotion is an AI-powered video-processing platform that can extract clips from source footage and generate merged video outputs. The work centred on the backend: building robust, scalable processing pipelines behind a job queue.
My Role & Contributions
I built the complete backend:
- Video processing pipelines — clip extraction and merged-video generation.
- Queue management — job orchestration with BullMQ on Redis.
- Distributed rendering — rendering work spread across workers for scale.
- AI clip-extraction workflows — automated selection and extraction of clips.
- Video-generation automation — programmatic output with Remotion and FFmpeg.
Key Features
- Clip extraction from source video.
- Merged video generation.
- Distributed rendering for throughput.
- Queue-based processing so heavy jobs run reliably.
Tech Stack & Architecture
- Backend: Node.js.
- Processing: FFmpeg and Remotion for programmatic video.
- Queue: BullMQ on Redis for job management and distributed rendering.
Challenges & Solutions
- Rendering at scale — video rendering is heavy, so the system distributes jobs across workers and manages them through a queue.
- Reliable pipelines — combining FFmpeg and Remotion into automated, repeatable clip-extraction and generation workflows.
Timeline
February 2026 – May 2026 · 4 months
Links
No public link is available for this project.