We are seeking an experienced Full-Stack Engineer / Backend Automation Specialist to build and deploy a high-volume, self-hosted, programmatic video generation and automated social media publishing system.
The engine will support a minimum of 100 individual members, generating and publishing **4-**6 unique videos per day per member (2 TikTok, 2 Facebook Reels, 2 Instagram Reels) for a total production load of **400-**600 unique videos rendered and scheduled daily.
To optimize infrastructure costs, the core rendering pipeline must be deployed on our own server or dedicated VPS utilizing self-hosted/open-source tools where applicable.
- Implement a self-hosted rendering engine (leveraging tools like ShotTower, ShotStack API, FFmpeg, or lightweight containerized dynamic rendering node clusters).
- Programmatically combine dynamic dynamic elements into 100% unique MP4 files: clip assets, visual overlays, audio/voiceover, burnt-in captions, dynamic text overlays, and Member ID watermark/tracking tags.
- Maintain strict queue management and asset processing pipelines to ensure low system load on a dedicated VPS.
- Build or configure an orchestration layer (using n8n self-hosted) to handle trigger logic, dynamic asset pairing, API calls, and scheduled queue distribution.
Variable Scheduling: Implement randomized/varied posting schedules across all 100 members so content is distributed naturally throughout the day rather than in uniform batches.
- Implement secure OAuth2 authentication flow allowing 100 members to connect their social accounts (TikTok, Facebook Page/Reels, Instagram Graph API).
- Handle token refresh cycles, error handling, and webhooks for automatic re-authentication notifications if access tokens expire.
- Programmatically schedule and publish videos directly to member profiles via official social media APIs.
- Setup local/S3 temporary storage for raw assets and rendered video output.
Implement an automated cron job/cleanup engine to hard-delete generated video files exactly 48 hours after successful publication to preserve disk space and keep hosting overhead low.
- Architecture Setup & VPS Deployment: Server configuration, containerization (Docker), security setup, and API configuration for n8n, ShotTower/ShotStack, and local database.
- Template & Pipeline Build: Dynamic dynamic video generation pipeline capable of handling unique text/clip/caption/Member ID injections.
- User Authentication Portal / Integration: Functional authorization workflow for members to attach TikTok, Meta (FB/IG) accounts.
- Testing Phase: End-to-end stress testing with test accounts to verify pipeline throughput (600 renders/day), OAuth resilience, scheduling variance, and automatic storage cleanup.
- Documentation & Handover: Clear deployment guide and operational documentation for self-hosting maintenance.
To be considered for this project, please provide the following in your bid:
- Relevant Experience: Brief details/portfolio of past projects involving programmatic video generation (FFmpeg, ShotStack, Remotion, MoviePy), self-hosted n8n workflows, or bulk social media API automation.
- Proposed Architecture: Brief overview of how you plan to structure the rendering engine on a standalone VPS to avoid CPU bottlenecks (e.g., ShotTower vs. cloud rendering vs. local FFmpeg queues).
- Fixed Price Proposal: Total fixed price for complete build, deployment, and testing.
- Estimated Lead Time: Complete timeline breakdown from milestone kickoff to final testing and handover.