Content Storyteller cover

AI/MLFeatured

Content Storyteller

Multimodal AI Platform for Marketing Assets

Source on GitHub (opens in new tab)

The problem

Marketing teams need polished multi-format output (copy, visuals, video, voiceover) from rough inputs — but most AI tools lock you into a single modality.

My approach

Output-intent inference decides which pipeline stages run per job, so the same brief yields a LinkedIn GIF, a storyboard, and a Veo promo from ten coordinated GCP services.

What I built

Transforms rough inputs (text, images, voice notes) into polished marketing assets — copy, visuals, storyboards, voiceover scripts, and short promo videos. Includes Trend Analyzer for AI-powered trend discovery across Instagram Reels, X/Twitter, and LinkedIn. Built on GCP with Vertex AI Gemini.

How it works

Output-intent inference determines which pipeline stages run per job. GIF generation for LinkedIn, live agent voice assistant with function calling and native audio, ZIP export bundle. 10 GCP services including Cloud Run, Pub/Sub, Firestore, Vertex AI.

What I'd do differently

Multimodal pipelines are cheaper when you infer intent upfront instead of running every stage blindly.

Good questions to ask

3 questions an interviewer could ask about Content Storyteller

  1. Why did you decide to infer intent upfront in your multimodal pipelines instead of running every stage blindly?

    Based on: Multimodal pipelines are cheaper when you infer intent upfront instead of running every stage blindly.

  2. How did you design output-intent inference to decide which pipeline stages run per job?

    Based on: Output-intent inference decides which pipeline stages run per job, so the same brief yields a LinkedIn GIF, a storyboard, and a Veo promo from ten coordinated GCP services.

  3. Why did you build this tool given that most AI tools lock you into a single modality?

    Based on: Marketing teams need polished multi-format output (copy, visuals, video, voiceover) from rough inputs — but most AI tools lock you into a single modality.

AI-written from this site's content

AI-generated · may be wrongHow it worksPrivacy

At a glance

Technologies
7
Language
TypeScript
Updated
Mar 2026
  • React
  • Vertex AI
  • Gemini 2.5 Flash/Pro
  • Cloud Run
  • Pub/Sub
  • Firestore
  • Terraform
  • #marketing
  • #multimodal
  • #vertex-ai
  • #gcp

Compare with another project

Related work

AI-generated · may be wrongRanked by meaning with Gemini embeddings of each write-up. Semantic is cosine similarity; 1 would be identical.How it worksPrivacy