Viewing enriched Twitter post
Practical Techniques for Context Engineering 💡 This is a fantastic blog post from @tuanacelik and @LoganMarkewich on a comprehensive breakdown of the types of context an LLM can interact with, and the core dimensions you have to consider: 1️⃣ Knowledge Base or tool selection -… https://t.co/r6XEtsJpio
{
"media": [
{
"type": "image",
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1940852245450608646/media_0.jpg?",
"filename": "media_0.jpg"
}
],
"nlp": {
"sentiment": "positive",
"topics": [
"context engineering",
"LLM interaction",
"blog post"
],
"entities": [
"tuanacelik",
"LoganMarkewich"
],
"summary": "The tweet praises a blog post by @tuanacelik and @LoganMarkewich about context engineering for LLMs.",
"language": "en",
"processed_at": "2025-08-07T09:56:40.037537"
}
}{
"data": {
"threaded_conversation_with_injections_v2": {
"instructions": [
{
"type": "TimelineClearCache"
},
{
"type": "TimelineAddEntries",
"entries": [
{
"entryId": "tweet-1940852245450608646",
"sortIndex": "1953485439260164096",
"content": {
"entryType": "TimelineTimelineItem",
"__typename": "TimelineTimelineItem",
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1940852245450608646",
"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": 64136,
"friends_count": 1378,
"has_custom_timelines": true,
"is_translator": false,
"listed_count": 1295,
"location": "",
"media_count": 1244,
"name": "Jerry Liu",
"normal_followers_count": 64136,
"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": [
"1940852245450608646"
],
"editable_until_msecs": "1751573809000",
"is_edit_eligible": true,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "29690",
"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": "Tm90ZVR3ZWV0OjE5NDA4NTIyNDUyNDUxODYwNDg=",
"text": "Practical Techniques for Context Engineering 💡\n\nThis is a fantastic blog post from @tuanacelik and @LoganMarkewich on a comprehensive breakdown of the types of context an LLM can interact with, and the core dimensions you have to consider: \n\n1️⃣ Knowledge Base or tool selection - the LLM needs information about what data sources are available in the first place\n2️⃣ Context Compression - this is where either retrieval or structured outputs or summarization comes into handy. Once you have the context you need to make sure you leave enough space for other things! \n3️⃣ Long-term Memory - there's ways you need to design auxiliary storage that can contain information which overflows the existing context, including vector based retrieval and fact retrieval. \n4️⃣ Workflow engineering - I talked a little about this yesterday, but this is how you orchestrate an entire sequence of calls to tackle the task, beyond context for a single step. \n\nCheck it out! https://t.co/hdfPIUuMEc",
"entity_set": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "llamaindex.ai/blog/context-e…",
"expanded_url": "https://www.llamaindex.ai/blog/context-engineering-what-it-is-and-techniques-to-consider?utm_source=socials&utm_medium=li_social",
"url": "https://t.co/hdfPIUuMEc",
"indices": [
959,
982
]
}
],
"user_mentions": [
{
"id_str": "209600624",
"name": "Tuana",
"screen_name": "tuanacelik",
"indices": [
83,
94
]
},
{
"id_str": "1676011773810360320",
"name": "Logan Markewich",
"screen_name": "LoganMarkewich",
"indices": [
99,
114
]
}
]
},
"richtext": {
"richtext_tags": [
{
"from_index": 412,
"to_index": 422,
"richtext_types": [
"Italic"
]
},
{
"from_index": 425,
"to_index": 444,
"richtext_types": [
"Italic"
]
},
{
"from_index": 447,
"to_index": 461,
"richtext_types": [
"Italic"
]
}
]
},
"media": {
"inline_media": []
}
}
}
},
"quoted_status_result": {
"result": {
"__typename": "Tweet",
"rest_id": "1940810514227196236",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxNjA0Mjc4MzU4Mjk2MDU1ODA4",
"rest_id": "1604278358296055808",
"affiliates_highlighted_label": {},
"has_graduated_access": true,
"is_blue_verified": true,
"profile_image_shape": "Circle",
"legacy": {
"can_dm": false,
"can_media_tag": true,
"created_at": "Sun Dec 18 00:52:44 +0000 2022",
"default_profile": true,
"default_profile_image": false,
"description": "Build AI agents over your documents\n\nGithub: https://t.co/HC19j7vMwc\nDocs: https://t.co/QInqg2zksh\nLlamaCloud: https://t.co/yQGTiRSNvj",
"entities": {
"description": {
"urls": [
{
"display_url": "github.com/run-llama/llam…",
"expanded_url": "http://github.com/run-llama/llama_index",
"url": "https://t.co/HC19j7vMwc",
"indices": [
45,
68
]
},
{
"display_url": "docs.llamaindex.ai",
"expanded_url": "http://docs.llamaindex.ai",
"url": "https://t.co/QInqg2zksh",
"indices": [
75,
98
]
},
{
"display_url": "cloud.llamaindex.ai",
"expanded_url": "https://cloud.llamaindex.ai/",
"url": "https://t.co/yQGTiRSNvj",
"indices": [
111,
134
]
}
]
},
"url": {
"urls": [
{
"display_url": "llamaindex.ai",
"expanded_url": "https://www.llamaindex.ai/",
"url": "https://t.co/epzefqQqZx",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 1425,
"followers_count": 99663,
"friends_count": 28,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 1571,
"location": "",
"media_count": 1676,
"name": "LlamaIndex 🦙",
"normal_followers_count": 99663,
"pinned_tweet_ids_str": [
"1950285220663742516"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1604278358296055808/1752258343",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1623505166996742144/n-PNQGgd_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "llama_index",
"statuses_count": 3500,
"translator_type": "none",
"url": "https://t.co/epzefqQqZx",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1940810514227196236"
],
"editable_until_msecs": "1751563859000",
"is_edit_eligible": true,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "40160",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://typefully.com/\" rel=\"nofollow\">Typefully</a>",
"note_tweet": {
"is_expandable": true,
"note_tweet_results": {
"result": {
"id": "Tm90ZVR3ZWV0OjE5NDA4MTA1MTQwMTMzNDM3NDc=",
"text": "While the term “prompt engineering” focused on providing the right instructions to an LLM, “context engineering” puts a lot more focus on filling the context window of an LLM with the most relevant information, wherever that information may come from.\n\n🧠 Context could include system prompts, chat history, retrieved knowledge, tool definitions, structured outputs, and global state - all competing for limited context window space\n⚙️ Key techniques involve smart knowledge base selection, strategic long-term memory storage, and structured information extraction and more\n🔧 LlamaIndex lets you break complex tasks into focused steps, preventing context overload through event-driven orchestration\n📊 On the other hand, LlamaExtract lets you convert complex documents into structured outputs that provide condensed, relevant context without crowding your prompts\n\nRead our deep dive into context engineering principles and implementation strategies: https://t.co/fE6cuzDJMj",
"entity_set": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "llamaindex.ai/blog/context-e…",
"expanded_url": "https://www.llamaindex.ai/blog/context-engineering-what-it-is-and-techniques-to-consider?utm_source=socials&utm_medium=li_social",
"url": "https://t.co/fE6cuzDJMj",
"indices": [
949,
972
]
}
],
"user_mentions": []
}
}
}
},
"legacy": {
"bookmark_count": 172,
"bookmarked": false,
"created_at": "Thu Jul 03 16:30:59 +0000 2025",
"conversation_id_str": "1940810514227196236",
"display_text_range": [
0,
277
],
"entities": {
"hashtags": [],
"media": [
{
"display_url": "pic.x.com/fR3fLWjdea",
"expanded_url": "https://x.com/llama_index/status/1940810514227196236/photo/1",
"id_str": "1940810511295434753",
"indices": [
278,
301
],
"media_key": "3_1940810511295434753",
"media_url_https": "https://pbs.twimg.com/media/Gu8k4I4b0AE_m6H.png",
"type": "photo",
"url": "https://t.co/fR3fLWjdea",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 379,
"w": 734,
"resize": "fit"
},
"medium": {
"h": 379,
"w": 734,
"resize": "fit"
},
"small": {
"h": 351,
"w": 680,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 379,
"width": 734,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 677,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 379,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 332,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 190,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 734,
"h": 379
}
]
},
"media_results": {
"result": {
"media_key": "3_1940810511295434753"
}
}
}
],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": []
},
"extended_entities": {
"media": [
{
"display_url": "pic.x.com/fR3fLWjdea",
"expanded_url": "https://x.com/llama_index/status/1940810514227196236/photo/1",
"id_str": "1940810511295434753",
"indices": [
278,
301
],
"media_key": "3_1940810511295434753",
"media_url_https": "https://pbs.twimg.com/media/Gu8k4I4b0AE_m6H.png",
"type": "photo",
"url": "https://t.co/fR3fLWjdea",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 379,
"w": 734,
"resize": "fit"
},
"medium": {
"h": 379,
"w": 734,
"resize": "fit"
},
"small": {
"h": 351,
"w": 680,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 379,
"width": 734,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 677,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 379,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 332,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 190,
"h": 379
},
{
"x": 0,
"y": 0,
"w": 734,
"h": 379
}
]
},
"media_results": {
"result": {
"media_key": "3_1940810511295434753"
}
}
}
]
},
"favorite_count": 161,
"favorited": false,
"full_text": "While the term “prompt engineering” focused on providing the right instructions to an LLM, “context engineering” puts a lot more focus on filling the context window of an LLM with the most relevant information, wherever that information may come from.\n\n🧠 Context could include… https://t.co/fR3fLWjdea",
"is_quote_status": false,
"lang": "en",
"possibly_sensitive": false,
"possibly_sensitive_editable": true,
"quote_count": 2,
"reply_count": 2,
"retweet_count": 29,
"retweeted": false,
"user_id_str": "1604278358296055808",
"id_str": "1940810514227196236"
}
}
},
"legacy": {
"bookmark_count": 378,
"bookmarked": false,
"created_at": "Thu Jul 03 19:16:49 +0000 2025",
"conversation_id_str": "1940852245450608646",
"display_text_range": [
0,
281
],
"entities": {
"hashtags": [],
"media": [
{
"display_url": "pic.x.com/r6XEtsJpio",
"expanded_url": "https://x.com/jerryjliu0/status/1940852245450608646/photo/1",
"id_str": "1940852112042479618",
"indices": [
282,
305
],
"media_key": "3_1940852112042479618",
"media_url_https": "https://pbs.twimg.com/media/Gu9Ktnvb0AIuFAs.jpg",
"type": "photo",
"url": "https://t.co/r6XEtsJpio",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 866,
"w": 1668,
"resize": "fit"
},
"medium": {
"h": 623,
"w": 1200,
"resize": "fit"
},
"small": {
"h": 353,
"w": 680,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 866,
"width": 1668,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 1546,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 866,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 760,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 433,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 1668,
"h": 866
}
]
},
"media_results": {
"result": {
"media_key": "3_1940852112042479618"
}
}
}
],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "209600624",
"name": "Tuana",
"screen_name": "tuanacelik",
"indices": [
83,
94
]
},
{
"id_str": "1676011773810360320",
"name": "Logan Markewich",
"screen_name": "LoganMarkewich",
"indices": [
99,
114
]
}
]
},
"extended_entities": {
"media": [
{
"display_url": "pic.x.com/r6XEtsJpio",
"expanded_url": "https://x.com/jerryjliu0/status/1940852245450608646/photo/1",
"id_str": "1940852112042479618",
"indices": [
282,
305
],
"media_key": "3_1940852112042479618",
"media_url_https": "https://pbs.twimg.com/media/Gu9Ktnvb0AIuFAs.jpg",
"type": "photo",
"url": "https://t.co/r6XEtsJpio",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 866,
"w": 1668,
"resize": "fit"
},
"medium": {
"h": 623,
"w": 1200,
"resize": "fit"
},
"small": {
"h": 353,
"w": 680,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 866,
"width": 1668,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 1546,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 866,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 760,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 433,
"h": 866
},
{
"x": 0,
"y": 0,
"w": 1668,
"h": 866
}
]
},
"media_results": {
"result": {
"media_key": "3_1940852112042479618"
}
}
}
]
},
"favorite_count": 274,
"favorited": false,
"full_text": "Practical Techniques for Context Engineering 💡\n\nThis is a fantastic blog post from @tuanacelik and @LoganMarkewich on a comprehensive breakdown of the types of context an LLM can interact with, and the core dimensions you have to consider: \n\n1️⃣ Knowledge Base or tool selection -… https://t.co/r6XEtsJpio",
"is_quote_status": true,
"lang": "en",
"possibly_sensitive": false,
"possibly_sensitive_editable": true,
"quote_count": 1,
"quoted_status_id_str": "1940810514227196236",
"quoted_status_permalink": {
"url": "https://t.co/KECddvzlTN",
"expanded": "https://twitter.com/llama_index/status/1940810514227196236",
"display": "x.com/llama_index/st…"
},
"reply_count": 0,
"retweet_count": 62,
"retweeted": false,
"user_id_str": "369777416",
"id_str": "1940852245450608646"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "Tweet"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet"
}
}
},
{
"entryId": "cursor-bottom-1953485439260164095",
"sortIndex": "1953485439260164095",
"content": {
"entryType": "TimelineTimelineCursor",
"__typename": "TimelineTimelineCursor",
"value": "DAAFCgABGxwsqDD___0LAAIAAAAoRW1QQzZ3QUFBZlEvZ0dKTjB2R3AvQUFBQUFFYTcwclZpNXBRQmc9PQgAAwAAAAIAAA",
"cursorType": "Bottom"
}
}
]
},
{
"type": "TimelineTerminateTimeline",
"direction": "Top"
},
{
"type": "TimelineTerminateTimeline",
"direction": "Bottom"
}
],
"metadata": {
"scribeConfig": {
"page": "ranked_replies"
}
}
}
}
}