Skip to main content
Back to challenges
Challenge

Develop a Generative AI Micro-Content Studio

This challenge tasks you with building an AI-powered micro-content studio. This studio will be an agentic system designed to rapidly generate engaging, brand-aligned short-form video concepts, scripts, and storyboards. You'll leverage GPT-5 for creative ideation and advanced scriptwriting, and Mixtral 8x22B for generating variations and personalization. The core will involve DSPy for programmatic prompting and quality control, orchestrated by Semantic Kernel, using MCP-enabled RAG against brand guidelines and product catalogs to ensure brand safety and product integration. Focus on iterative generation and adaptive thinking to meet creative constraints.

Special Purpose AgentsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Create an agentic micro-content studio using GPT-5 and DSPy, orchestrated by Semantic Kernel, for generating brand-aligned shoppable video concepts with MCP-enabled RAG.

Delivery guide

How work is evaluated

Evaluation

The evaluation will assess the micro-content studio's ability to generate creative, brand-compliant, and shoppable micro-content concepts (script + storyboard), demonstrating effective use of DSPy, Semantic Kernel orchestration, and MCP-enabled RAG.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Learning goals

What you should walk away with

  • Utilize MCP-enabled RAG against brand guidelines and product catalogs for brand safety and product integration.

  • Orchestrate agent workflows using Semantic Kernel with GPT-5 and Mixtral 8x22B.

  • Implement DSPy for optimizing LLM prompts and building generative pipelines.

  • Design an agentic micro-content studio for generating short-form video concepts.

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 1 tasks

Challenge setup

The evaluation will assess the micro-content studio's ability to generate creative, brand-compliant, and shoppable micro-content concepts (script + storyboard), demonstrating effective use of DSPy, Semantic Kernel orc...

Challenge input

{'product_focus': {'name': 'Luxury Smartwatch', 'features': ['health tracking', 'stylish design'], 'price': '$499'}, 'brand_guidelines': {'tone': '...

Agent execution

No agent tool is listed for this challenge.

Evaluated output

{'concept_title': '...', 'logline': '...', 'script': {'scene_1': '...', 'scene_2': '...'}, 'storyboard_description': [{'scene': 1, 'visuals': '...'...

Evaluation checks
  • All required output fields (script, storyboard, etc.) are present and logically structured.
  • The product is clearly and naturally integrated into the content concept.
  • The content's tone and visual style align with the provided brand guidelines.
Configured evaluation evidence
  • CreativityAndEngagement target: 0.85
Configured runtime
  • Python execution harness

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • All required output fields (script, storyboard, etc.) are present and logically structured.
  • The product is clearly and naturally integrated into the content concept.
  • The content's tone and visual style align with the provided brand guidelines.

Metrics

  • CreativityAndEngagement · Target 0.85
  • DSPyPipelineEfficiency · Target 0.8

Workflow tasks

  1. Task 1

    Input: {'product_focus': {'name': 'Luxury Smartwatch', 'features': ['health tracking', 'stylish design'], 'price': '$499'}, 'brand_guidelines': {'tone': 'aspirational, modern', 'visual_style': 'clean, minimalist', 'call_to_action': 'Shop now at [BrandWebsite]'}, 'audience_segment': 'young professionals'}

    Output: {'concept_title': '...', 'logline': '...', 'script': {'scene_1': '...', 'scene_2': '...'}, 'storyboard_description': [{'scene': 1, 'visuals': '...', 'audio': '...'}], 'product_integration_points': ['...'], 'brand_compliance_score': 0.0, 'dspy_trace': '...'}

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Develop a Generative AI Micro-Content Studio