Viewing enriched Twitter post
“Is RAG dead?” This question pops up every 2 weeks! I just got the scoop from @HamelHusain and @sh_reya in their amazing AI Evals course: https://t.co/hSpuJqh6eM Let’s clear the confusion—and talk about what actually matters when evaluating retrieval-augmented generation… https://t.co/o2P17tZYX3
{
"media": [
{
"type": "image",
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1929972685355864295/media_0.jpg?",
"filename": "media_0.jpg"
}
],
"nlp": {
"sentiment": "neutral",
"topics": [
"retrieval-augmented generation",
"AI Evals course",
"evaluation methods"
],
"entities": [
"Hamel Husain",
"sh_reya"
],
"summary": "The tweet discusses the recurring question about the relevance of retrieval-augmented generation and highlights insights from an AI Evals course.",
"language": "en",
"processed_at": "2025-08-07T11:30:49.272503"
}
}{
"data": {
"threaded_conversation_with_injections_v2": {
"instructions": [
{
"type": "TimelineClearCache"
},
{
"type": "TimelineAddEntries",
"entries": [
{
"entryId": "tweet-1929972685355864295",
"sortIndex": "1953509132191399936",
"content": {
"entryType": "TimelineTimelineItem",
"__typename": "TimelineTimelineItem",
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929972685355864295",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxNjgzMDI3OTAwNDIzODg4ODk2",
"rest_id": "1683027900423888896",
"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 Jul 23 08:15:04 +0000 2023",
"default_profile": true,
"default_profile_image": false,
"description": "Founder\n🤖🧠 Epistemic Evals\n🛠️ AI Alignment. Building in public, open source \nhttps://t.co/5y8EvhoqPI \nhttps://t.co/3QTFaXT5mb",
"entities": {
"description": {
"urls": [
{
"display_url": "epistemicme.ai",
"expanded_url": "http://epistemicme.ai",
"url": "https://t.co/5y8EvhoqPI",
"indices": [
77,
100
]
},
{
"display_url": "github.com/Epistemic-Me/",
"expanded_url": "https://github.com/Epistemic-Me/",
"url": "https://t.co/3QTFaXT5mb",
"indices": [
102,
125
]
}
]
},
"url": {
"urls": [
{
"display_url": "linktr.ee/therobertta",
"expanded_url": "https://linktr.ee/therobertta",
"url": "https://t.co/dm4Fo3NyZE",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 200,
"followers_count": 120,
"friends_count": 211,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 0,
"location": "The Universe",
"media_count": 76,
"name": "Robert Ta",
"normal_followers_count": 120,
"pinned_tweet_ids_str": [
"1933257514231214363"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1683027900423888896/1746046732",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1944967824490807297/0tVLMIEU_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "therobertta_",
"statuses_count": 909,
"translator_type": "none",
"url": "https://t.co/dm4Fo3NyZE",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"professional": {
"rest_id": "1827733725771399598",
"professional_type": "Creator",
"category": [
{
"id": 713,
"name": "Science & Technology",
"icon_name": ""
}
]
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929972685355864295"
],
"editable_until_msecs": "1748979920000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "3075",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://buffer.com\" rel=\"nofollow\">Buffer</a>",
"note_tweet": {
"is_expandable": true,
"note_tweet_results": {
"result": {
"id": "Tm90ZVR3ZWV0OjE5Mjk5NzI2ODUyMDkwMjI0NjU=",
"text": "“Is RAG dead?” This question pops up every 2 weeks!\n\nI just got the scoop from @HamelHusain and @sh_reya in their amazing AI Evals course: https://t.co/HMPRuJfn5Q \n\nLet’s clear the confusion—and talk about what actually matters when evaluating retrieval-augmented generation pipelines. 🧵",
"entity_set": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "maven.com/parlance-labs/…",
"expanded_url": "https://maven.com/parlance-labs/evals",
"url": "https://t.co/HMPRuJfn5Q",
"indices": [
139,
162
]
}
],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
79,
91
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
96,
104
]
}
]
}
}
}
},
"legacy": {
"bookmark_count": 7,
"bookmarked": false,
"created_at": "Tue Jun 03 18:45:20 +0000 2025",
"conversation_id_str": "1929972685355864295",
"display_text_range": [
0,
275
],
"entities": {
"hashtags": [],
"media": [
{
"display_url": "pic.x.com/o2P17tZYX3",
"expanded_url": "https://x.com/therobertta_/status/1929972685355864295/photo/1",
"id_str": "1929972682558173184",
"indices": [
276,
299
],
"media_key": "3_1929972682558173184",
"media_url_https": "https://pbs.twimg.com/media/Gsij7eoWcAA5x-_.jpg",
"type": "photo",
"url": "https://t.co/o2P17tZYX3",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 1500,
"w": 1200,
"resize": "fit"
},
"medium": {
"h": 1200,
"w": 960,
"resize": "fit"
},
"small": {
"h": 680,
"w": 544,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 1500,
"width": 1200,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 1200,
"h": 672
},
{
"x": 0,
"y": 0,
"w": 1200,
"h": 1200
},
{
"x": 0,
"y": 0,
"w": 1200,
"h": 1368
},
{
"x": 225,
"y": 0,
"w": 750,
"h": 1500
},
{
"x": 0,
"y": 0,
"w": 1200,
"h": 1500
}
]
},
"media_results": {
"result": {
"media_key": "3_1929972682558173184"
}
}
}
],
"symbols": [],
"timestamps": [],
"urls": [
{
"display_url": "maven.com/parlance-labs/…",
"expanded_url": "https://maven.com/parlance-labs/evals",
"url": "https://t.co/hSpuJqh6eM",
"indices": [
139,
162
]
}
],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
79,
91
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
96,
104
]
}
]
},
"extended_entities": {
"media": [
{
"display_url": "pic.x.com/o2P17tZYX3",
"expanded_url": "https://x.com/therobertta_/status/1929972685355864295/photo/1",
"id_str": "1929972682558173184",
"indices": [
276,
299
],
"media_key": "3_1929972682558173184",
"media_url_https": "https://pbs.twimg.com/media/Gsij7eoWcAA5x-_.jpg",
"type": "photo",
"url": "https://t.co/o2P17tZYX3",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"medium": {
"faces": []
},
"small": {
"faces": []
},
"orig": {
"faces": []
}
},
"sizes": {
"large": {
"h": 1500,
"w": 1200,
"resize": "fit"
},
"medium": {
"h": 1200,
"w": 960,
"resize": "fit"
},
"small": {
"h": 680,
"w": 544,
"resize": "fit"
},
"thumb": {
"h": 150,
"w": 150,
"resize": "crop"
}
},
"original_info": {
"height": 1500,
"width": 1200,
"focus_rects": [
{
"x": 0,
"y": 0,
"w": 1200,
"h": 672
},
{
"x": 0,
"y": 0,
"w": 1200,
"h": 1200
},
{
"x": 0,
"y": 0,
"w": 1200,
"h": 1368
},
{
"x": 225,
"y": 0,
"w": 750,
"h": 1500
},
{
"x": 0,
"y": 0,
"w": 1200,
"h": 1500
}
]
},
"media_results": {
"result": {
"media_key": "3_1929972682558173184"
}
}
}
]
},
"favorite_count": 4,
"favorited": false,
"full_text": "“Is RAG dead?” This question pops up every 2 weeks!\n\nI just got the scoop from @HamelHusain and @sh_reya in their amazing AI Evals course: https://t.co/hSpuJqh6eM \n\nLet’s clear the confusion—and talk about what actually matters when evaluating retrieval-augmented generation… https://t.co/o2P17tZYX3",
"is_quote_status": false,
"lang": "en",
"possibly_sensitive": false,
"possibly_sensitive_editable": true,
"quote_count": 1,
"reply_count": 1,
"retweet_count": 3,
"retweeted": false,
"user_id_str": "1683027900423888896",
"id_str": "1929972685355864295"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "SelfThread"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet"
}
}
},
{
"entryId": "conversationthread-1929972686895128857",
"sortIndex": "1953509132191399926",
"content": {
"entryType": "TimelineTimelineModule",
"__typename": "TimelineTimelineModule",
"items": [
{
"entryId": "conversationthread-1929972686895128857-tweet-1929972686895128857",
"item": {
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929972686895128857",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxNjgzMDI3OTAwNDIzODg4ODk2",
"rest_id": "1683027900423888896",
"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 Jul 23 08:15:04 +0000 2023",
"default_profile": true,
"default_profile_image": false,
"description": "Founder\n🤖🧠 Epistemic Evals\n🛠️ AI Alignment. Building in public, open source \nhttps://t.co/5y8EvhoqPI \nhttps://t.co/3QTFaXT5mb",
"entities": {
"description": {
"urls": [
{
"display_url": "epistemicme.ai",
"expanded_url": "http://epistemicme.ai",
"url": "https://t.co/5y8EvhoqPI",
"indices": [
77,
100
]
},
{
"display_url": "github.com/Epistemic-Me/",
"expanded_url": "https://github.com/Epistemic-Me/",
"url": "https://t.co/3QTFaXT5mb",
"indices": [
102,
125
]
}
]
},
"url": {
"urls": [
{
"display_url": "linktr.ee/therobertta",
"expanded_url": "https://linktr.ee/therobertta",
"url": "https://t.co/dm4Fo3NyZE",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 200,
"followers_count": 120,
"friends_count": 211,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 0,
"location": "The Universe",
"media_count": 76,
"name": "Robert Ta",
"normal_followers_count": 120,
"pinned_tweet_ids_str": [
"1933257514231214363"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1683027900423888896/1746046732",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1944967824490807297/0tVLMIEU_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "therobertta_",
"statuses_count": 909,
"translator_type": "none",
"url": "https://t.co/dm4Fo3NyZE",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"professional": {
"rest_id": "1827733725771399598",
"professional_type": "Creator",
"category": [
{
"id": 713,
"name": "Science & Technology",
"icon_name": ""
}
]
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929972686895128857"
],
"editable_until_msecs": "1748979920000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "95",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://buffer.com\" rel=\"nofollow\">Buffer</a>",
"legacy": {
"bookmark_count": 0,
"bookmarked": false,
"created_at": "Tue Jun 03 18:45:20 +0000 2025",
"conversation_id_str": "1929972685355864295",
"display_text_range": [
22,
249
],
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
0,
12
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
13,
21
]
}
]
},
"favorite_count": 0,
"favorited": false,
"full_text": "@HamelHusain @sh_reya (1/)\n\nRAG = Retrieval-Augmented Generation\n\nVanilla RAG = simple fetch → feed to LLM\n\nAgentic RAG = more steps, more looping, more decisions\n\nYet at its core, it's still...\n\nUser Query → Retrieval → Generation\n\nRAG is not dead.",
"in_reply_to_screen_name": "therobertta_",
"in_reply_to_status_id_str": "1929972685355864295",
"in_reply_to_user_id_str": "1683027900423888896",
"is_quote_status": false,
"lang": "en",
"quote_count": 0,
"reply_count": 1,
"retweet_count": 0,
"retweeted": false,
"user_id_str": "1683027900423888896",
"id_str": "1929972686895128857"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "SelfThread"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
},
{
"entryId": "conversationthread-1929972686895128857-tweet-1929972687750725634",
"item": {
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929972687750725634",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxNjgzMDI3OTAwNDIzODg4ODk2",
"rest_id": "1683027900423888896",
"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 Jul 23 08:15:04 +0000 2023",
"default_profile": true,
"default_profile_image": false,
"description": "Founder\n🤖🧠 Epistemic Evals\n🛠️ AI Alignment. Building in public, open source \nhttps://t.co/5y8EvhoqPI \nhttps://t.co/3QTFaXT5mb",
"entities": {
"description": {
"urls": [
{
"display_url": "epistemicme.ai",
"expanded_url": "http://epistemicme.ai",
"url": "https://t.co/5y8EvhoqPI",
"indices": [
77,
100
]
},
{
"display_url": "github.com/Epistemic-Me/",
"expanded_url": "https://github.com/Epistemic-Me/",
"url": "https://t.co/3QTFaXT5mb",
"indices": [
102,
125
]
}
]
},
"url": {
"urls": [
{
"display_url": "linktr.ee/therobertta",
"expanded_url": "https://linktr.ee/therobertta",
"url": "https://t.co/dm4Fo3NyZE",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 200,
"followers_count": 120,
"friends_count": 211,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 0,
"location": "The Universe",
"media_count": 76,
"name": "Robert Ta",
"normal_followers_count": 120,
"pinned_tweet_ids_str": [
"1933257514231214363"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1683027900423888896/1746046732",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1944967824490807297/0tVLMIEU_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "therobertta_",
"statuses_count": 909,
"translator_type": "none",
"url": "https://t.co/dm4Fo3NyZE",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"professional": {
"rest_id": "1827733725771399598",
"professional_type": "Creator",
"category": [
{
"id": 713,
"name": "Science & Technology",
"icon_name": ""
}
]
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929972687750725634"
],
"editable_until_msecs": "1748979920000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "105",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://buffer.com\" rel=\"nofollow\">Buffer</a>",
"legacy": {
"bookmark_count": 0,
"bookmarked": false,
"created_at": "Tue Jun 03 18:45:20 +0000 2025",
"conversation_id_str": "1929972685355864295",
"display_text_range": [
22,
259
],
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
0,
12
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
13,
21
]
}
]
},
"favorite_count": 0,
"favorited": false,
"full_text": "@HamelHusain @sh_reya (2/)\n\nWhat IS dead: pretending you don’t need to evaluate the retrieval layer.\n\nMost LLM hallucinations?\n\nGarbage in = garbage out.\n\nPut lipstick on a pig, it's still a pig.\n\nYou can’t debug a gen step if your retrieval missed key facts.",
"in_reply_to_screen_name": "therobertta_",
"in_reply_to_status_id_str": "1929972686895128857",
"in_reply_to_user_id_str": "1683027900423888896",
"is_quote_status": false,
"lang": "en",
"quote_count": 0,
"reply_count": 1,
"retweet_count": 0,
"retweeted": false,
"user_id_str": "1683027900423888896",
"id_str": "1929972687750725634"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "SelfThread"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
},
{
"entryId": "conversationthread-1929972686895128857-tweet-1929972688677654705",
"item": {
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929972688677654705",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxNjgzMDI3OTAwNDIzODg4ODk2",
"rest_id": "1683027900423888896",
"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 Jul 23 08:15:04 +0000 2023",
"default_profile": true,
"default_profile_image": false,
"description": "Founder\n🤖🧠 Epistemic Evals\n🛠️ AI Alignment. Building in public, open source \nhttps://t.co/5y8EvhoqPI \nhttps://t.co/3QTFaXT5mb",
"entities": {
"description": {
"urls": [
{
"display_url": "epistemicme.ai",
"expanded_url": "http://epistemicme.ai",
"url": "https://t.co/5y8EvhoqPI",
"indices": [
77,
100
]
},
{
"display_url": "github.com/Epistemic-Me/",
"expanded_url": "https://github.com/Epistemic-Me/",
"url": "https://t.co/3QTFaXT5mb",
"indices": [
102,
125
]
}
]
},
"url": {
"urls": [
{
"display_url": "linktr.ee/therobertta",
"expanded_url": "https://linktr.ee/therobertta",
"url": "https://t.co/dm4Fo3NyZE",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 200,
"followers_count": 120,
"friends_count": 211,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 0,
"location": "The Universe",
"media_count": 76,
"name": "Robert Ta",
"normal_followers_count": 120,
"pinned_tweet_ids_str": [
"1933257514231214363"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1683027900423888896/1746046732",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1944967824490807297/0tVLMIEU_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "therobertta_",
"statuses_count": 909,
"translator_type": "none",
"url": "https://t.co/dm4Fo3NyZE",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"professional": {
"rest_id": "1827733725771399598",
"professional_type": "Creator",
"category": [
{
"id": 713,
"name": "Science & Technology",
"icon_name": ""
}
]
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929972688677654705"
],
"editable_until_msecs": "1748979920000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "106",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://buffer.com\" rel=\"nofollow\">Buffer</a>",
"legacy": {
"bookmark_count": 2,
"bookmarked": false,
"created_at": "Tue Jun 03 18:45:20 +0000 2025",
"conversation_id_str": "1929972685355864295",
"display_text_range": [
22,
293
],
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
0,
12
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
13,
21
]
}
]
},
"favorite_count": 0,
"favorited": false,
"full_text": "@HamelHusain @sh_reya (3/)\n\nThe fix...\n\nEvaluate retrieval and generation separately.\n\nRecall@K\n\nPrecision@K\n\nMRR (Mean Reciprocal Rank)\n\nNDCG (rank-aware relevance).... etc.\n\nEach tells you a different story. \n\nPick based on your app. \n\nBut the key is evaluate each of those steps separately.",
"in_reply_to_screen_name": "therobertta_",
"in_reply_to_status_id_str": "1929972687750725634",
"in_reply_to_user_id_str": "1683027900423888896",
"is_quote_status": false,
"lang": "en",
"quote_count": 0,
"reply_count": 1,
"retweet_count": 0,
"retweeted": false,
"user_id_str": "1683027900423888896",
"id_str": "1929972688677654705"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "SelfThread"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
},
{
"entryId": "conversationthread-1929972686895128857-tweet-1929972689654927401",
"item": {
"itemContent": {
"itemType": "TimelineTweet",
"__typename": "TimelineTweet",
"tweet_results": {
"result": {
"__typename": "Tweet",
"rest_id": "1929972689654927401",
"has_birdwatch_notes": false,
"core": {
"user_results": {
"result": {
"__typename": "User",
"id": "VXNlcjoxNjgzMDI3OTAwNDIzODg4ODk2",
"rest_id": "1683027900423888896",
"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 Jul 23 08:15:04 +0000 2023",
"default_profile": true,
"default_profile_image": false,
"description": "Founder\n🤖🧠 Epistemic Evals\n🛠️ AI Alignment. Building in public, open source \nhttps://t.co/5y8EvhoqPI \nhttps://t.co/3QTFaXT5mb",
"entities": {
"description": {
"urls": [
{
"display_url": "epistemicme.ai",
"expanded_url": "http://epistemicme.ai",
"url": "https://t.co/5y8EvhoqPI",
"indices": [
77,
100
]
},
{
"display_url": "github.com/Epistemic-Me/",
"expanded_url": "https://github.com/Epistemic-Me/",
"url": "https://t.co/3QTFaXT5mb",
"indices": [
102,
125
]
}
]
},
"url": {
"urls": [
{
"display_url": "linktr.ee/therobertta",
"expanded_url": "https://linktr.ee/therobertta",
"url": "https://t.co/dm4Fo3NyZE",
"indices": [
0,
23
]
}
]
}
},
"fast_followers_count": 0,
"favourites_count": 200,
"followers_count": 120,
"friends_count": 211,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 0,
"location": "The Universe",
"media_count": 76,
"name": "Robert Ta",
"normal_followers_count": 120,
"pinned_tweet_ids_str": [
"1933257514231214363"
],
"possibly_sensitive": false,
"profile_banner_url": "https://pbs.twimg.com/profile_banners/1683027900423888896/1746046732",
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1944967824490807297/0tVLMIEU_normal.jpg",
"profile_interstitial_type": "",
"screen_name": "therobertta_",
"statuses_count": 909,
"translator_type": "none",
"url": "https://t.co/dm4Fo3NyZE",
"verified": false,
"want_retweets": false,
"withheld_in_countries": []
},
"professional": {
"rest_id": "1827733725771399598",
"professional_type": "Creator",
"category": [
{
"id": 713,
"name": "Science & Technology",
"icon_name": ""
}
]
},
"tipjar_settings": {}
}
}
},
"unmention_data": {},
"edit_control": {
"edit_tweet_ids": [
"1929972689654927401"
],
"editable_until_msecs": "1748979921000",
"is_edit_eligible": false,
"edits_remaining": "5"
},
"is_translatable": false,
"views": {
"count": "85",
"state": "EnabledWithCount"
},
"source": "<a href=\"https://buffer.com\" rel=\"nofollow\">Buffer</a>",
"note_tweet": {
"is_expandable": true,
"note_tweet_results": {
"result": {
"id": "Tm90ZVR3ZWV0OjE5Mjk5NzI2ODk1NjY4MzQ2ODg=",
"text": "(4/)\n\nMy takeaway learnings from class today...\n\nDon’t get distracted by shiny agentic wrappers.\nDon’t ignore your retriever’s blind spots.\nBuild synthetic queries grounded in real user behavior.\n\nGet the retrieval right—like REALLY right. \n\nGo look at the actual FAQs or pieces of content. Dig in.\n\nKnow it well.\n\nAwesome course by @HamelHusain and @sh_reya. \n\nHighly recommend to ANYONE trying to level up their AI game. 💪 \n\n👉 I'll write more about my learnings as I take the course, stay tuned!",
"entity_set": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
333,
345
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
350,
358
]
}
]
}
}
}
},
"legacy": {
"bookmark_count": 0,
"bookmarked": false,
"created_at": "Tue Jun 03 18:45:21 +0000 2025",
"conversation_id_str": "1929972685355864295",
"display_text_range": [
22,
301
],
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": [
{
"id_str": "825766640",
"name": "Hamel Husain",
"screen_name": "HamelHusain",
"indices": [
0,
12
]
},
{
"id_str": "2286218053",
"name": "Shreya Shankar",
"screen_name": "sh_reya",
"indices": [
13,
21
]
}
]
},
"favorite_count": 0,
"favorited": false,
"full_text": "@HamelHusain @sh_reya (4/)\n\nMy takeaway learnings from class today...\n\nDon’t get distracted by shiny agentic wrappers.\nDon’t ignore your retriever’s blind spots.\nBuild synthetic queries grounded in real user behavior.\n\nGet the retrieval right—like REALLY right. \n\nGo look at the actual FAQs or pieces…",
"in_reply_to_screen_name": "therobertta_",
"in_reply_to_status_id_str": "1929972688677654705",
"in_reply_to_user_id_str": "1683027900423888896",
"is_quote_status": false,
"lang": "en",
"quote_count": 0,
"reply_count": 0,
"retweet_count": 0,
"retweeted": false,
"user_id_str": "1683027900423888896",
"id_str": "1929972689654927401"
},
"quick_promote_eligibility": {
"eligibility": "IneligibleNotProfessional"
}
}
},
"tweetDisplayType": "SelfThread"
},
"clientEventInfo": {
"component": "tweet",
"element": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
}
],
"metadata": {
"conversationMetadata": {
"allTweetIds": [
"1929972689654927401"
],
"enableDeduplication": true
}
},
"displayType": "VerticalConversation",
"clientEventInfo": {
"component": "tweet",
"details": {
"conversationDetails": {
"conversationSection": "HighQuality"
},
"timelinesDetails": {
"controllerData": "DAACDAAEDAABCgABAAAAAAAAAAEKAAIAAAAAAAAAAAAAAAA="
}
}
}
}
},
{
"entryId": "cursor-bottom-1953509132191399925",
"sortIndex": "1953509132191399925",
"content": {
"entryType": "TimelineTimelineCursor",
"__typename": "TimelineTimelineCursor",
"value": "DAAFCgABGxxCNKG___MLAAIAAABQRW1QQzZ3QUFBZlEvZ0dKTjB2R3AvQUFBQUFVYXlLUHVrTmZRNXhySW8rN3NseUVaR3Npajc1RVdVQ2theUtQdlZ0WlFzUnJJbys4ZmxvQUMIAAMAAAACAAA",
"cursorType": "Bottom"
}
}
]
},
{
"type": "TimelineTerminateTimeline",
"direction": "Top"
},
{
"type": "TimelineTerminateTimeline",
"direction": "Bottom"
}
],
"metadata": {
"reader_mode_config": {
"is_reader_mode_available": true
},
"scribeConfig": {
"page": "ranked_replies"
}
}
}
}
}