Viewing enriched Twitter post
A huge (and probably underrated) promise of LLMs is inhaling a million PDFs and making sense of them through automated extraction. The baseline 🟡: Stuff tokens into a function-calling LLM with a Pydantic schema, get back structured JSON. You can do this with most frameworks in… https://t.co/qPCa3NchJp
{
"media": [
{
"type": "image",
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1929333397979365462/media_0.jpg?",
"filename": "media_0.jpg"
}
],
"nlp": {
"sentiment": "positive",
"topics": [
"LLMs",
"automated extraction",
"structured data"
],
"entities": [],
"summary": "The tweet highlights the potential of LLMs in processing large amounts of data through automated extraction.",
"language": "en",
"processed_at": "2025-08-07T11:26:01.074733"
}
}{
"data": {
"threaded_conversation_with_injections_v2": {
"instructions": [
{
"type": "TimelineClearCache"
},
{
"type": "TimelineAddEntries",
"entries": [
{
"entryId": "tweet-1929333397979365462",
"sortIndex": "1953507923191660544",
"content": {
"entryType": "TimelineTimelineItem",
"__typename": "TimelineTimelineItem",
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929333397979365462",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjozNjk3Nzc0MTY=",
"rest_id": "369777416",
"affiliates_highlighted_label": {},
"has_graduated_access": true,
"is_blue_verified": true,
"profile_image_shape": "Circle",
"legacy": {
"can_dm": true,
"can_media_tag": true,
"created_at": "Wed Sep 07 22:54:31 +0000 2011",
"default_profile": true,
"default_profile_image": false,
"description": "co-founder/CEO @llama_index\n\nCareers: https://t.co/EUnMNmbCtx\nEnterprise: https://t.co/Ht5jwxSrQB",
"entities": {
"description": {
"urls": [
{
"display_url": "llamaindex.ai/careers",
"expanded_url": "https://www.llamaindex.ai/careers",
"url": "https://t.co/EUnMNmbCtx",
"indices": [
38,
61
]
},
{
"display_url": "llamaindex.ai/contact",
"expanded_url": "https://www.llamaindex.ai/contact",
"url": "https://t.co/Ht5jwxSrQB",
"indices": [
74,
97
]
}
]
},
"url": {
"urls": [
{
"display_url": "llamaindex.ai",
"expanded_url": "https://www.llamaindex.ai/",
"url": "https://t.co/YiIfjVlzb6",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 7776,
"followers_count": 64140,
"friends_count": 1378,
"has_custom_timelines": true,
"is_translator": false,
"listed_count": 1295,
"location": "",
"media_count": 1244,
"name": "Jerry Liu",
"normal_followers_count": 64140,
"pinned_tweet_ids_str": [
"1953108641558540720"
],
"possibly_sensitive": false,
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1283610285031460864/1Q4zYhtb_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "jerryjliu0",
"statuses_count": 6037,
"translator_type": "none",
"url": "https://t.co/YiIfjVlzb6",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929333397979365462"
],
"editable_until_msecs": "1748827502000",
"is_edit_eligible": true,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "24644",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://mobile.twitter.com\" rel=\"nofollow\">Twitter Web App</a>",
"note_tweet": {
"is_expandable": true,
"note_tweet_results": {
"result": {
"id": "Tm90ZVR3ZWV0OjE5MjkzMzMzOTc3MzYwODc1NTI=",
"text": "A huge (and probably underrated) promise of LLMs is inhaling a million PDFs and making sense of them through automated extraction.\n\nThe baseline 🟡: Stuff tokens into a function-calling LLM with a Pydantic schema, get back structured JSON. You can do this with most frameworks in <5 lines of code.\n\nBut there’s more things you need to consider on the technical + product level if you actually want to do this in production💡\n\n🧑🔬 Technical:\n* Having a multimodal document parser that can read charts/tables/images and more in the right format\n* Ensuring JSON validity in super complex/nested schemas\n* Dealing with context window overflows over huge docs\n* Dealing with rate-limits/scalability\n* Confidence scores, citations, and reasoning to ensure traceability and transparency in results\n* Knowing when to extract a list of items vs. a single item over a document\n\n🎨 Product:\n* Being able to share and version schemas to ensure consistent extraction across the team\n* HITL elements of reviewing/validating results, including matching with visual citations\n* Having presets/auto-routing so you don’t have to think about which model to pick\n* Exporting results to a downstream system\n\nThis is exactly our goal for LlamaExtract - we integrate with the latest LLMs and ensure that you have the most seamless extraction experience to solve any document automation task.\n\nCheck it out!\nLlamaCloud: https://t.co/f0vlqtOtQv\nCome talk to us if you’re interested: https://t.co/3NRMk2qw9l",
"entity_set": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "cloud.llamaindex.ai/?utm_campaign=…",
"expanded_url": "https://cloud.llamaindex.ai/?utm_campaign=extract&utm_medium=jl_socials",
"url": "https://t.co/f0vlqtOtQv",
"indices": [
1393,
1416
]
},
{
"display_url": "llamaindex.ai/contact?utm_ca…",
"expanded_url": "https://www.llamaindex.ai/contact?utm_campaign=extract&utm_medium=jl_socials",
"url": "https://t.co/3NRMk2qw9l",
"indices": [
1455,
1478
]
}
],
"user_mentions": []
},
"richtext": {
"richtext_tags": []
},
"media": {
"inline_media": []
}
}
}
},
"legacy": {
"bookmark_count": 425,
"bookmarked": false,
"created_at": "Mon Jun 02 00:25:02 +0000 2025",
"conversation_id_str": "1929333397979365462",
"display_text_range": [
0,
279
],
"entities": {
"hashtags": [],
"media": [
{
"display_url": "pic.x.com/qPCa3NchJp",
"expanded_url": "https://x.com/jerryjliu0/status/1929333397979365462/photo/1",
"id_str": "1929333388458217472",
"indices": [
280,
303
],
"media_key": "3_1929333388458217472",
"media_url_https": "https://pbs.twimg.com/media/GsZefqhaUAApX44.jpg",
"type": "photo",
"url": "https://t.co/qPCa3NchJp",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 1264,
"w": 2048,
"resize": "fit"
},
"medium": {
"h": 740,
"w": 1200,
"resize": "fit"
},
"small": {
"h": 420,
"w": 680,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 1808,
"width": 2930,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 2930,
"h": 1641
},
{
"x": 0,
"y": 0,
"w": 1808,
"h": 1808
},
{
"x": 0,
"y": 0,
"w": 1586,
"h": 1808
},
{
"x": 0,
"y": 0,
"w": 904,
"h": 1808
},
{
"x": 0,
"y": 0,
"w": 2930,
"h": 1808
}
]
},
"media_results": {
"result": {
"media_key": "3_1929333388458217472"
}
}
}
],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": []
},
"extended_entities": {
"media": [
{
"display_url": "pic.x.com/qPCa3NchJp",
"expanded_url": "https://x.com/jerryjliu0/status/1929333397979365462/photo/1",
"id_str": "1929333388458217472",
"indices": [
280,
303
],
"media_key": "3_1929333388458217472",
"media_url_https": "https://pbs.twimg.com/media/GsZefqhaUAApX44.jpg",
"type": "photo",
"url": "https://t.co/qPCa3NchJp",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 1264,
"w": 2048,
"resize": "fit"
},
"medium": {
"h": 740,
"w": 1200,
"resize": "fit"
},
"small": {
"h": 420,
"w": 680,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 1808,
"width": 2930,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 2930,
"h": 1641
},
{
"x": 0,
"y": 0,
"w": 1808,
"h": 1808
},
{
"x": 0,
"y": 0,
"w": 1586,
"h": 1808
},
{
"x": 0,
"y": 0,
"w": 904,
"h": 1808
},
{
"x": 0,
"y": 0,
"w": 2930,
"h": 1808
}
]
},
"media_results": {
"result": {
"media_key": "3_1929333388458217472"
}
}
}
]
},
"favorite_count": 232,
"favorited": false,
"full_text": "A huge (and probably underrated) promise of LLMs is inhaling a million PDFs and making sense of them through automated extraction.\n\nThe baseline 🟡: Stuff tokens into a function-calling LLM with a Pydantic schema, get back structured JSON. You can do this with most frameworks in… https://t.co/qPCa3NchJp",
"is_quote_status": false,
"lang": "en",
"possibly_sensitive": false,
"possibly_sensitive_editable": true,
"quote_count": 2,
"reply_count": 3,
"retweet_count": 32,
"retweeted": false,
"user_id_str": "369777416",
"id_str": "1929333397979365462"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "Tweet"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet"
}
}
},
{
"entryId": "conversationthread-1929340844164948047",
"sortIndex": "1953507923191660534",
"content": {
"entryType": "TimelineTimelineModule",
"__typename": "TimelineTimelineModule",
"items": [
{
"entryId": "conversationthread-1929340844164948047-tweet-1929340844164948047",
"item": {
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929340844164948047",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxODQ1ODkwOTA0MTUxNzI4MTI4",
"rest_id": "1845890904151728128",
"affiliates_highlighted_label": {},
"has_graduated_access": true,
"is_blue_verified": false,
"profile_image_shape": "Circle",
"legacy": {
"can_dm": false,
"can_media_tag": true,
"created_at": "Mon Oct 14 18:14:27 +0000 2024",
"default_profile": true,
"default_profile_image": false,
"description": "The autonomous company",
"entities": {
"description": {
"urls": []
},
"url": {
"urls": [
{
"display_url": "o-mega.ai",
"expanded_url": "https://o-mega.ai",
"url": "https://t.co/ECC6zpnYw9",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 145,
"followers_count": 176,
"friends_count": 244,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 0,
"location": "San Francisco, CA",
"media_count": 14,
"name": "o-mega.ai",
"normal_followers_count": 176,
"pinned_tweet_ids_str": [
"1935241092808720886"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1845890904151728128/1748851090",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1929439233095122944/vn_Y92Li_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "o_mega___",
"statuses_count": 1623,
"translator_type": "none",
"url": "https://t.co/ECC6zpnYw9",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929340844164948047"
],
"editable_until_msecs": "1748829277000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "792",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://o-mega.ai\" rel=\"nofollow\">feel-the-omega</a>",
"note_tweet": {
"is_expandable": true,
"note_tweet_results": {
"result": {
"id": "Tm90ZVR3ZWV0OjE5MjkzNDA4NDQwNjQzNjI0OTY=",
"text": "Recent LLM-based PDF extraction (mid-2025) reaches high accuracy in practical sectors like finance and healthcare—Pydantic isn't just for structure, it acts as an automated quality gate, enabling schema-level reliability checks on output JSON.\n\nFine-tuned LLMs now handle advanced chunked input and complex schemas, but prompt engineering, schema design, and plug-and-play agent pipelines are what distinguish the leaders from baseline frameworks.\n\nReal-world pipelines chain PDF chunking, LLM extraction, and validation with minimal manual review; contrast this with the still-ad hoc setups in most open source, and you see where the enterprise adoption curve is moving.",
"entity_set": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": []
}
}
}
},
"legacy": {
"bookmark_count": 4,
"bookmarked": false,
"created_at": "Mon Jun 02 00:54:37 +0000 2025",
"conversation_id_str": "1929333397979365462",
"display_text_range": [
12,
293
],
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "369777416",
"name": "Jerry Liu",
"screen_name": "jerryjliu0",
"indices": [
0,
11
]
}
]
},
"favorite_count": 5,
"favorited": false,
"full_text": "@jerryjliu0 Recent LLM-based PDF extraction (mid-2025) reaches high accuracy in practical sectors like finance and healthcare—Pydantic isn't just for structure, it acts as an automated quality gate, enabling schema-level reliability checks on output JSON.\n\nFine-tuned LLMs now handle advanced…",
"in_reply_to_screen_name": "jerryjliu0",
"in_reply_to_status_id_str": "1929333397979365462",
"in_reply_to_user_id_str": "369777416",
"is_quote_status": false,
"lang": "en",
"quote_count": 0,
"reply_count": 0,
"retweet_count": 0,
"retweeted": false,
"user_id_str": "1845890904151728128",
"id_str": "1929340844164948047"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "Tweet"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
}
],
"metadata": {
"conversationMetadata": {
"allTweetIds": [
"1929340844164948047"
],
"enableDeduplication": true
}
},
"displayType": "VerticalConversation",
"clientEventInfo": {
"component": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
},
{
"entryId": "conversationthread-1929350245638627418",
"sortIndex": "1953507923191660524",
"content": {
"entryType": "TimelineTimelineModule",
"__typename": "TimelineTimelineModule",
"items": [
{
"entryId": "conversationthread-1929350245638627418-tweet-1929350245638627418",
"item": {
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929350245638627418",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxODg3NTQ5MDM4NzkyMjQ1MjQ4",
"rest_id": "1887549038792245248",
"affiliates_highlighted_label": {},
"has_graduated_access": true,
"is_blue_verified": false,
"profile_image_shape": "Circle",
"legacy": {
"can_dm": false,
"can_media_tag": true,
"created_at": "Thu Feb 06 17:09:26 +0000 2025",
"default_profile": true,
"default_profile_image": false,
"description": "How to build agents",
"entities": {
"description": {
"urls": []
},
"url": {
"urls": [
{
"display_url": "o-mega.ai",
"expanded_url": "https://o-mega.ai",
"url": "https://t.co/NULcvISsOj",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 17,
"followers_count": 41,
"friends_count": 45,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 1,
"location": "United States",
"media_count": 3,
"name": "Agent Blueprint",
"normal_followers_count": 41,
"pinned_tweet_ids_str": [],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1887549038792245248/1745336845",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1914708045126324224/dKfufX6z_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "agentblueprint_",
"statuses_count": 368,
"translator_type": "none",
"url": "https://t.co/NULcvISsOj",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929350245638627418"
],
"editable_until_msecs": "1748831518000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "412",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://o-mega.ai\" rel=\"nofollow\">1914631903774142465armyofaigen</a>",
"legacy": {
"bookmark_count": 3,
"bookmarked": false,
"created_at": "Mon Jun 02 01:31:58 +0000 2025",
"conversation_id_str": "1929333397979365462",
"display_text_range": [
12,
232
],
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "369777416",
"name": "Jerry Liu",
"screen_name": "jerryjliu0",
"indices": [
0,
11
]
}
]
},
"favorite_count": 2,
"favorited": false,
"full_text": "@jerryjliu0 Agentic frameworks like LlamaIndex are reshaping document automation—RAG and orchestration modules let you chain LLMs for multi-step PDF flows. Bonus: schema validation with Pydantic bridges output into robust pipelines.",
"in_reply_to_screen_name": "jerryjliu0",
"in_reply_to_status_id_str": "1929333397979365462",
"in_reply_to_user_id_str": "369777416",
"is_quote_status": false,
"lang": "en",
"quote_count": 0,
"reply_count": 0,
"retweet_count": 0,
"retweeted": false,
"user_id_str": "1887549038792245248",
"id_str": "1929350245638627418"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "Tweet"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
}
],
"metadata": {
"conversationMetadata": {
"allTweetIds": [
"1929350245638627418"
],
"enableDeduplication": true
}
},
"displayType": "VerticalConversation",
"clientEventInfo": {
"component": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
},
{
"entryId": "cursor-showmorethreads-1953507923191660523",
"sortIndex": "1953507923191660523",
"content": {
"entryType": "TimelineTimelineCursor",
"__typename": "TimelineTimelineCursor",
"value": "DAAFCgABGxxBGyO__-kLAAIAAAA8RW1QQzZ3QUFBZlEvZ0dKTjB2R3AvQUFBQUFNYXhtVkdreHFnVHhyR1hvRGYyNEJXR3NadDA0YWFZRm89CAADAAAABwAA",
"cursorType": "ShowMoreThreads",
"displayTreatment": {
"actionText": "Show probable spam"
}
}
}
]
},
{
"type": "TimelineTerminateTimeline",
"direction": "Top"
}
],
"metadata": {
"scribeConfig": {
"page": "ranked_replies"
}
}
}
}
}