Owais Abdullah logo
Back to all projects
AI ToolFree for Personal UseAttribution Required

SocialFTE

Autonomous AI Social Media Employee (Digital FTE) with Video Rendering & Human-in-the-Loop

Free & Open Architecture Alternative To:

HootsuiteBufferOpus ClipPredis.aiLately AI

💰 Value & Cost Savings Comparison

Traditional Industry Solution

Hootsuite Enterprise + Opus Clip + Video Editor

$600 - $1,500 / month

This Project (By Owais Abdullah)

Free for Personal Use (Self-Hosted Architecture)

Save $10,000+/year on social media management tools

🎯 Pain Points Solved

Why this project was built and the operational friction it eliminates:

The Problem

Managing consistent daily social media posts across 4 platforms requires 2–3 dedicated employees (copywriter, video editor, and community manager).

The Solution

SocialFTE handles the full lifecycle autonomously: writing captions, generating graphics, rendering MP4 video shorts, and scheduling posts.

The Problem

Automated social bots post generic AI sludge with em dashes, robotic hype, and off-brand phrasing that harms customer trust.

The Solution

Enforces strict brand guidelines (real prices upfront, no em dashes, conversational tone) with zero tolerance for generic AI fluff.

The Problem

Fully autonomous posting is risky; a single hallucinated post or incorrect price can spark public relations issues.

The Solution

Rigid human-in-the-loop governance: posts MUST be approved via Discord interactive button cards before touching live platform APIs.

The Problem

Creating video content programmatically is historically difficult and brittle with legacy video stitching scripts.

The Solution

Leverages Remotion (React/TypeScript) to render pixel-perfect, typography-driven animated vertical videos and YouTube shorts on demand.

Core Capabilities & Architecture

Split into two primary services: apps/dashboard (Next.js 16, Tailwind CSS, Neon Postgres) providing the visual calendar and review UI, and apps/worker (Python 3.12, LiteLLM, OpenAI Agents SDK) executing background drafting, media processing, and publishing. Video compositions live in packages/remotion. Approved posts trigger GitHub Actions dispatch or local headless workers to render MP4s, upload to Cloudflare R2, and distribute via platform APIs.

  • Dual Architecture: High-performance Next.js 16 Dashboard UI + Python background AI worker
  • Remotion Video Pipeline: Renders programmatic MP4 videos with dynamic captions, audio mixing, and motion graphics
  • Discord Interactive Approvals: Posts are delivered as rich cards with 'Approve', 'Edit', or 'Reject' buttons
  • Multi-Platform Publishing: Direct API publishing to Facebook, Instagram, YouTube Shorts, and TikTok
  • Vision-Scored Cover Frame Picker: Evaluates best video frames for thumbnail click-through rate optimization
  • Audit Trail & Rescheduling: Interactive calendar with daily-cap warnings and full execution logging
Built With:Next.js 16Python 3.12+Remotion (TSX)OpenAI Agents SDKCloudflare R2Discord APIFFmpeg & WhisperNeon PostgreSQL

🛠️ How to Setup & Run Locally

Quick-start guide to clone, configure, and execute this project in your own environment:

Prerequisites

  • Node.js 20+ and pnpm / npm
  • Python 3.12+ with uv
  • FFmpeg installed on your system
  • Discord Bot Token and Webhook URL
  • Cloudflare R2 bucket for asset storage

Key Environment Variables

VariableDescriptionExample
OPENROUTER_API_KEYLLM API key for caption draftingsk-or-v1-...
DISCORD_BOT_TOKENBot token for sending approval cardsMTIz...
DISCORD_CHANNEL_IDChannel ID where drafts are routed1234567890
DATABASE_URLPostgreSQL connection string (Neon pgvector)postgresql://...
R2_ACCESS_KEY_IDCloudflare R2 access key for media storagecf_key_...

Execution Steps

1.Clone repository

git clone https://github.com/MrOwaisAbdullah/socialfte.git
cd socialfte

2.Install frontend & Remotion dependencies

npm install

3.Install Python worker dependencies

cd apps/worker && uv sync && cd ../..

4.Launch with Docker Compose or locally

docker compose up -d

Starts the dashboard on http://localhost:3000 and the background worker daemon.

🗺️ Project Roadmap & Planned Improvements

Current release progress and planned future enhancements:

ElevenLabs Voiceover Auto-Sync

Integrate synthetic voiceover generation aligned word-for-word with Remotion kinetic typography.

In Progress

Direct TikTok API Audit

Graduate TikTok integration from draft-only mode to full automated publishing upon developer review.

In Progress

Engagement Analytics Feedback Loop

Fetch view, like, and comment metrics to train agent prompts on top-performing post angles.

Planned

License, Fair Use & Attribution Terms

Creative Commons Attribution-NonCommercial 4.0 (CC BY-NC 4.0)

Free for personal, educational, and showcase use with mandatory credit to Owais Abdullah. Commercial deployment for clients or SaaS resale requires a separate license.

Required Attribution Notice: "Built by Owais Abdullah (https://github.com/MrOwaisAbdullah)"