AI-Powered Personalized News Shorts Generator
This challenge involves building an agentic system to automate the generation of personalized, short-form news content suitable for a vertical video feed. You will use AutoGen to orchestrate a team of agents that leverage OpenAI o3's multimodal capabilities for understanding source content and generating concise summaries/scripts, and Claude Sonnet 4 for content moderation and style adaptation. The system should dynamically tailor content based on simulated user profiles and preferences, demonstrating cutting-edge content transformation and personalization.
What you are building
The core problem, expected build, and operating context for this challenge.
This challenge involves building an agentic system to automate the generation of personalized, short-form news content suitable for a vertical video feed. You will use AutoGen to orchestrate a team of agents that leverage OpenAI o3's multimodal capabilities for understanding source content and generating concise summaries/scripts, and Claude Sonnet 4 for content moderation and style adaptation. The system should dynamically tailor content based on simulated user profiles and preferences, demonstrating cutting-edge content transformation and personalization.
Shared data for this challenge
Review public datasets and any private uploads tied to your build.
What you should walk away with
Master AutoGen for orchestrating dynamic, multi-agent conversations and workflows to manage complex content generation pipelines, from source ingestion to final personalized output.
Leverage OpenAI o3's advanced multimodal capabilities for understanding news content across text articles, audio transcripts, and video descriptions, and generating engaging, concise summaries or scripts optimized for short-form vertical video.
Design and implement personalization algorithms that dynamically adapt content style, length, focus, and tone based on simulated user profiles (e.g., 'tech enthusiast', 'policy wonk', 'brief overview seeker') and past interaction data.
Integrate Claude Sonnet 4 for nuanced content moderation, ensuring generated summaries adhere to ethical guidelines, maintain journalistic integrity, avoid bias, and adapt to specific editorial style guides.
Develop tools for AutoGen agents to access real-time news feeds (RSS, News APIs), extract key information, and generate basic visual cues or descriptions suitable for enhancing short-form content generation.
Implement adaptive thinking mechanisms where agents adjust their summarization strategy (e.g., focus on key facts, emotional impact, or specific angles) based on the target user persona and the detected complexity of the news story.
How this agent runs
The system will be evaluated on the relevance, conciseness, and personalization quality of its generated news shorts for given user profiles and source articles. Accuracy and adherence to content guidelines will also...
Challenge input
{'news_article_url': 'string', 'user_profile': {'name': 'string', 'interests': ['string'], 'preferred_style': 'string', 'max_length_words': 'intege...
Agent execution
The configured agent processes the input under the challenge policy.
Evaluated output
{'summary_text': 'string', 'reading_time_seconds': 'integer', 'style_adherence_score': 'float', 'personalization_notes': 'string'}
- Ensure the generated summary is grammatically correct and logically flows.
- Verify the summary adheres to the 'max_length_words' specified in the user profile (allowing for a small bu...
- Personalization Score target: 4
- Python execution harness
View technical recipe
Configured tools
No tool records are attached.
Evaluation contract
- The evaluation module defines the checks.
Recipe state
This is a preview. The configuration can change before the evaluation recipe is locked.
Run this agent on your dataset and AI stack
Bring your dataset, model providers, and success criteria. We will scope the right managed run for your team.
Scope a managed run[ok] Wrote CHALLENGE.md
[ok] Wrote .versalist.json
[ok] Wrote eval/examples.json
Requires VERSALIST_API_KEY. Works with any MCP-aware editor.
DocsFind another challenge
Jump to a random challenge when you want a fresh benchmark or a different problem space.