AI-Assisted Creative for Automated Marketing using Python
summary
A production pipeline written mostly in Python that automatically renders thousands of platform-specific video and image assets via Adobe After Effects rendering, pulls images from the iChef CMS, and stages outputs in Dropbox to produce a CSV catalog consumed by Meta for paid social. The system runs end-to-end with no manual pre-release checks, completes new-show promotion within 24–48 hours, uses python's OpenAI API and rule checks for AI-assisted QA, and supports human intervention through Plotly dash apps and Slack alerts.
AI-written summary · about
chapters (15)
- 0:00AI-assisted marketing project
- 1:04Long-tail consumption in streaming
- 3:11Marketing assets at scale
- 6:27Generative fill limitations
- 7:50Dynamic creative explained
- 8:11Catalog ads for streaming
- 10:22Python and After Effects system
- 13:31Asset creation pipeline
- 14:59Catalog CSV for Meta
- 17:28Deploying to Meta
- 18:30System health and errors
- 20:35AI image detection
- 21:40Slack monitoring
- 23:02Human intervention tools
- 26:53System status and plans
related talks
30mCreating an AI-Python Infrastructure that works with a reasonable scale
14mAI-Video Ads – Built 90% Faster & 80% Cheaper
45mThe Role of AI in Modern Marketing
01h 43mAI as a Service
19mAI video is Already App-ening
14mAI's Impact on the Future of Marketing
16mAI no-code for marketing
21mFrom Workflows to Agents: The AI Shift Redefining UA Creative Production