Miadi Review

Automating Instagram Reels Generation with Remotion and AI

Sep 30, 2026 · model: manual-import · channel: Dispatch

Watch source

In the chronicle

Automating Instagram Reels Generation with Remotion and AI

Source: Remotion: Create Videos Programmatically! — Dispatch
https://youtu.be/DCeh1USebns STATUS: OUTDATED, we are starting to think about a team for production and one for 'production context exploration' which is not yet represented in jgwill/miadi-orchestration-kit teams / plugins but some tmux sessions can be surfed and a subsection under 'Intent' that is short could be expressed (like the Chronicle Staging/germination and the one that this review is meant to (creating means of creating videos for the Miadi TV Series inspired by mature miadi-chronicles)

Intent

  • Producer/supplier of video for the miadi-chronicle
  • Agent's team at a certain point produce a video of what they created (the transition from where we were when we started the work to now as well as what the next team will be capable todo with the work that our team completed

The two teams this review serves (stub, 2026-10-06)

  • T5 · Production context exploration (your term; its name is yours): concepts fed and explored before an episode exists, in the Chronicle staging circle. Lead seat: the staging session.
  • T6 · Production (name held, D11): makes video of what the teams created: where the work started, where it is now, and what the next team can do with it. This review's pipeline (Remotion) is a candidate instrument.

Both are drafted in jgwill/miadi-orchestration-kit https://github.com/jgwill/miadi-orchestration-kit/blob/main/teams/README.md. Written by Mino-Bimaadizi-Daa from your note on review 6c3f477f v10.

Summary

This engineering walkthrough demonstrates a fully automated pipeline designed to generate and publish 100 stock comparison Instagram Reels in one hour. By integrating Remotion, React, Chart.js, MySQL, Upstox API, OpenAI, FFmpeg, and Vercel Blob Storage, the system autonomously fetches financial data, renders synchronized animated graph videos, attaches audio, and uploads them directly to Instagram.

Aspects

Data Aggregation & AI-Driven Stock Matching (00:57-01:47)

The backend queries unprocessed stock records from a MySQL database and utilizes OpenAI completions to suggest two correlated stocks for multi-ticker performance comparisons. Upstox API supplies the historical price series required for chart generation.

Deterministic Graph Interpolation in Remotion (02:29-03:00)

To eliminate frame synchronization glitches between Remotion's frame tick and Chart.js animations, graph data points are manually interpolated per frame using useCurrentFrame and useVideoConfig hooks, guaranteeing deterministic visual line progression.

Media Packaging & Orchestration Pipeline (03:43-04:44)

A PowerShell orchestrator renders Remotion video assets, invokes FFmpeg to merge background audio tracks and extract final-frame thumbnails, and triggers a Python script to push assets to Vercel Blob and publish to Instagram.

End-results and Core Aspect

Core Aspect — End-to-End Autonomous Video Production Engine (00:32-01:25)

A headless multi-stack pipeline orchestrates AI selection, market data retrieval, programmatic React video rendering, audio-visual multiplexing, and social media posting without human intervention.

Results

  • High-Throughput Batch Publishing (05:01-05:16): Executing a looping automation script successfully generated and published 100 unique, fully styled stock comparison Reels to an active Instagram account within a single hour.
  • Seamless Audio & Cover Image Extraction (04:13-04:30): FFmpeg pipeline integration systematically binds background MP3 audio into the rendered MP4 video and extracts high-resolution thumbnail images from the final animation frame.
  • Glitch-Free Animated Line Visualizations (02:43-03:08): Mathematical per-frame data slice interpolation delivers smooth, synchronized 30 FPS multi-dataset financial charts within Remotion's isolated rendering environment.