🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

Outstanding paper on long-horizon agents. (bookmark it) Similar to humans, how do you make agents persist on a difficult task, and how is that useful? And which models today work well on this? This new work, AutoLab, explores this question and how encoding persistence in agents is beneficial for tasks such as auto research and engineering tasks. Can a model keep improving an artifact for hours, under a strict wall-clock budget, the way real research and engineering actually work? Results: AutoLab hands agents 36 expert-curated tasks across system optimization, model development, CUDA kernels, and puzzles, each starting from a correct but deliberately suboptimal baseline. Across 17 frontier models, the dominant predictor of success was not the quality of the first attempt. It was persistence, repeatedly benchmarking, editing, and folding in empirical feedback. It appears that Claude-opus-4.6 sustained that loop well. Most of the other models quit early or burned the budget, making almost no progress. Paper: https://t.co/jb8uYR0fpE Learn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c

Media 1

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2062570078705688777/media_0.jpg",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-06-16T11:17:15.535645",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2062570078705688777",
  "url": "https://x.com/dair_ai/status/2062570078705688777",
  "twitterUrl": "https://twitter.com/dair_ai/status/2062570078705688777",
  "text": "Outstanding paper on long-horizon agents.\n\n(bookmark it)\n\nSimilar to humans, how do you make agents persist on a difficult task, and how is that useful?\n\nAnd which models today work well on this?\n\nThis new work, AutoLab, explores this question and how encoding persistence in agents is beneficial for tasks such as auto research and engineering tasks.\n\nCan a model keep improving an artifact for hours, under a strict wall-clock budget, the way real research and engineering actually work?\n\nResults:\n\nAutoLab hands agents 36 expert-curated tasks across system optimization, model development, CUDA kernels, and puzzles, each starting from a correct but deliberately suboptimal baseline.\n\nAcross 17 frontier models, the dominant predictor of success was not the quality of the first attempt. It was persistence, repeatedly benchmarking, editing, and folding in empirical feedback.\n\nIt appears that Claude-opus-4.6 sustained that loop well. Most of the other models quit early or burned the budget, making almost no progress.\n\nPaper: https://t.co/jb8uYR0fpE\n\nLearn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c",
  "source": "Twitter for iPhone",
  "retweetCount": 47,
  "replyCount": 21,
  "likeCount": 276,
  "quoteCount": 2,
  "viewCount": 21101,
  "createdAt": "Thu Jun 04 16:20:02 +0000 2026",
  "lang": "en",
  "bookmarkCount": 332,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2062570078705688777",
  "displayTextRange": [
    0,
    275
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "dair_ai",
    "url": "https://x.com/dair_ai",
    "twitterUrl": "https://twitter.com/dair_ai",
    "id": "889050642903293953",
    "name": "DAIR.AI",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1643277398522187778/31dedbLo_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/889050642903293953/1773242460",
    "description": "Democratizing AI research, education, and technologies. Learn about AI Agents for FREE at https://t.co/HHXg8rryu4",
    "location": "",
    "followers": 126150,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": [
          {
            "display_url": "academy.dair.ai/courses/elemen…",
            "expanded_url": "https://academy.dair.ai/courses/elements-of-ai-agents",
            "indices": [
              90,
              113
            ],
            "url": "https://t.co/HHXg8rryu4"
          }
        ]
      },
      "url": {
        "urls": [
          {
            "display_url": "dair.ai",
            "expanded_url": "https://www.dair.ai/",
            "indices": [
              0,
              23
            ],
            "url": "https://t.co/lkqPZtMU5s"
          }
        ]
      }
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4593,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 245,
    "statusesCount": 3316,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2066548421670941068"
    ],
    "profile_bio": {},
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.x.com/Y4Y4vZvewF",
        "expanded_url": "https://x.com/dair_ai/status/2062570078705688777/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "medium": {
            "faces": []
          },
          "orig": {
            "faces": []
          },
          "small": {
            "faces": []
          }
        },
        "id_str": "2062570075073421315",
        "indices": [
          276,
          299
        ],
        "media_key": "3_2062570075073421315",
        "media_results": {
          "result": {
            "media_key": "3_2062570075073421315"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HJ-4jikaIAMR6aX.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 877,
              "w": 1566,
              "x": 0,
              "y": 0
            },
            {
              "h": 1566,
              "w": 1566,
              "x": 0,
              "y": 0
            },
            {
              "h": 1785,
              "w": 1566,
              "x": 0,
              "y": 0
            },
            {
              "h": 1898,
              "w": 949,
              "x": 0,
              "y": 0
            },
            {
              "h": 1898,
              "w": 1566,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1898,
          "width": 1566
        },
        "sizes": {
          "large": {
            "h": 1898,
            "resize": "fit",
            "w": 1566
          },
          "medium": {
            "h": 1200,
            "resize": "fit",
            "w": 990
          },
          "small": {
            "h": 680,
            "resize": "fit",
            "w": 561
          },
          "thumb": {
            "h": 150,
            "resize": "crop",
            "w": 150
          }
        },
        "type": "photo",
        "url": "https://t.co/Y4Y4vZvewF"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2606.05080",
        "expanded_url": "https://arxiv.org/abs/2606.05080",
        "indices": [
          1032,
          1055
        ],
        "url": "https://t.co/jb8uYR0fpE"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          1108,
          1131
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "article": null
}