🐦 Twitter Post Details

Viewing enriched Twitter post

@hamsabastani

🚨🚨 Excited to share our first *positive* results on AI in education! Most AI tutor work focuses on making the chatbot better. We suggest another lever: deciding what students should practice next to improve learning. We combine an LLM tutor with reinforcement learning to personalize problem sequencing using signals from student-chatbot interactions and solution attempts. We tested this in a 5-month randomized field experiment in a Python course across 10 high schools in Taipei. All students had the same course material and the same AI tutor. The only difference was adaptive vs. fixed problem sequencing. Result: across 770 students, adaptive sequencing improved performance on an in-person final exam taken without AI assistance by 0.15 SD, with larger effects for beginners. Our evidence suggests the gains came from stronger engagement and more productive AI use.

Media 1

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2033876571430248856/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2033876571430248856/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-03-17T13:45:35.719331",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2033876571430248856",
  "url": "https://x.com/hamsabastani/status/2033876571430248856",
  "twitterUrl": "https://twitter.com/hamsabastani/status/2033876571430248856",
  "text": "🚨🚨 Excited to share our first *positive* results on AI in education!\n\nMost AI tutor work focuses on making the chatbot better. We suggest another lever: deciding what students should practice next to improve learning.\n\nWe combine an LLM tutor with reinforcement learning to personalize problem sequencing using signals from student-chatbot interactions and solution attempts. \n\nWe tested this in a 5-month randomized field experiment in a Python course across 10 high schools in Taipei. All students had the same course material and the same AI tutor. The only difference was adaptive vs. fixed problem sequencing.\n\nResult: across 770 students, adaptive sequencing improved performance on an in-person final exam taken without AI assistance by 0.15 SD, with larger effects for beginners. Our evidence suggests the gains came from stronger engagement and more productive AI use.",
  "source": "Twitter for iPhone",
  "retweetCount": 2,
  "replyCount": 3,
  "likeCount": 16,
  "quoteCount": 1,
  "viewCount": 416,
  "createdAt": "Tue Mar 17 12:02:16 +0000 2026",
  "lang": "en",
  "bookmarkCount": 5,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2033876571430248856",
  "displayTextRange": [
    0,
    273
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "hamsabastani",
    "url": "https://x.com/hamsabastani",
    "twitterUrl": "https://twitter.com/hamsabastani",
    "id": "1112196703468302336",
    "name": "Hamsa Bastani",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1683176420153884673/GHWGBoFK_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/1112196703468302336/1655963165",
    "description": "",
    "location": "",
    "followers": 1733,
    "following": 405,
    "status": "",
    "canDm": false,
    "canMediaTag": true,
    "createdAt": "Sun Mar 31 03:35:32 +0000 2019",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 1043,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 27,
    "statusesCount": 404,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2033876571430248856"
    ],
    "profile_bio": {
      "description": "Prof @Wharton @Penn; machine learning for health & social good; foodie, gamer, homebody",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [],
          "user_mentions": [
            {
              "id_str": "0",
              "indices": [
                5,
                13
              ],
              "name": "",
              "screen_name": "Wharton"
            },
            {
              "id_str": "0",
              "indices": [
                14,
                19
              ],
              "name": "",
              "screen_name": "Penn"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "hamsabastani.github.io",
              "expanded_url": "https://hamsabastani.github.io/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/CJZ8g9ha7O"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/5xctGTywcS",
        "expanded_url": "https://twitter.com/hamsabastani/status/2033876571430248856/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2033874422566948868",
        "indices": [
          274,
          297
        ],
        "media_key": "3_2033874422566948868",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARw5xgLq2xAECgACHDnH9z0W4ZgAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHDnGAurbEAQKAAIcOcf3PRbhmAAA",
            "media_key": "3_2033874422566948868"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HDnGAurbEAQe_3G.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 752,
              "w": 1342,
              "x": 0,
              "y": 0
            },
            {
              "h": 972,
              "w": 972,
              "x": 0,
              "y": 0
            },
            {
              "h": 972,
              "w": 853,
              "x": 0,
              "y": 0
            },
            {
              "h": 972,
              "w": 486,
              "x": 0,
              "y": 0
            },
            {
              "h": 972,
              "w": 1342,
              "x": 0,
              "y": 0
            }
          ],
          "height": 972,
          "width": 1342
        },
        "sizes": {
          "large": {
            "h": 972,
            "w": 1342
          }
        },
        "type": "photo",
        "url": "https://t.co/5xctGTywcS"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}