Viewing enriched Twitter post
We’re excited to integrate @AskMarvinAI for automated metadata extraction from any text corpus. Simply add an annotation to a Pydantic model, and log metadata from *all* text chunks with that schema. Can use on its own or e2e for better RAG pipelines: https://t.co/jVLi64RB5G https://t.co/2CSW17jP9f
{
"media": [
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1697632035186372745/media_0.jpg",
"type": "photo",
"original_url": "https://pbs.twimg.com/media/F48zSBOawAAcXMc.jpg",
"download_date": "2025-08-13T06:03:01.611024",
"stored_in_supabase": true
},
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1697632035186372745/media_0.jpg",
"type": "photo",
"original_url": "https://pbs.twimg.com/media/F48zUHbbYAALA9r.jpg",
"download_date": "2025-08-13T06:03:02.694848",
"stored_in_supabase": true
}
],
"conversion_date": "2025-08-13T00:42:02.410475",
"format_converted": true,
"original_structure": "had_media_only"
}{
"user": {
"created_at": "2022-12-18T00:52:44.000Z",
"default_profile_image": false,
"description": "The data framework for LLMs\n\nGithub: https://t.co/Fvk4KTsVJY\nDiscord: https://t.co/3ktq3zzqTa\nDocs: https://t.co/coV2uzMQzr\nhttps://t.co/UXeIlwuXm2",
"fast_followers_count": 0,
"favourites_count": 395,
"followers_count": 29478,
"friends_count": 11,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 574,
"location": "",
"media_count": 290,
"name": "LlamaIndex 🦙",
"normal_followers_count": 29478,
"possibly_sensitive": false,
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1623505166996742144/n-PNQGgd_normal.jpg",
"screen_name": "llama_index",
"statuses_count": 973,
"translator_type": "none",
"url": "https://t.co/epzefqPT9Z",
"verified": false,
"withheld_in_countries": [],
"id_str": "1604278358296055808"
},
"id": "1697632035186372745",
"conversation_id": "1697632035186372745",
"full_text": "We’re excited to integrate @AskMarvinAI for automated metadata extraction from any text corpus.\n\nSimply add an annotation to a Pydantic model, and log metadata from *all* text chunks with that schema.\n\nCan use on its own or e2e for better RAG pipelines: https://t.co/jVLi64RB5G https://t.co/2CSW17jP9f",
"reply_count": 5,
"retweet_count": 17,
"favorite_count": 93,
"hashtags": [],
"symbols": [],
"user_mentions": [
{
"id_str": "1617539997497688067",
"name": "Marvin 🤖🧦",
"screen_name": "AskMarvinAI",
"profile": "https://twitter.com/AskMarvinAI"
}
],
"urls": [
{
"url": "https://t.co/jVLi64RB5G",
"expanded_url": "https://gpt-index.readthedocs.io/en/latest/examples/metadata_extraction/MarvinMetadataExtractorDemo.html",
"display_url": "gpt-index.readthedocs.io/en/latest/exam…"
}
],
"media": [
{
"media_url": "https://pbs.twimg.com/media/F48zSBOawAAcXMc.jpg",
"type": "photo"
},
{
"media_url": "https://pbs.twimg.com/media/F48zUHbbYAALA9r.jpg",
"type": "photo"
}
],
"url": "https://twitter.com/llama_index/status/1697632035186372745",
"created_at": "2023-09-01T15:26:32.000Z",
"#sort_index": "1697632035186372745",
"view_count": 8051,
"quote_count": 3,
"is_quote_tweet": false,
"is_retweet": false,
"is_pinned": false,
"is_truncated": false,
"startUrl": "https://twitter.com/llama_index/status/1697632035186372745"
}