🐦 Twitter Post Details

Viewing enriched Twitter post

@hwchase17

🎙️Introducing Max Agency Max Agency is a new podcast where we go deep on how the best agents are actually being built: architecture decisions, tradeoffs, evals, and everything in between. Each episode, I sit down with engineering leaders who are doing this work in production. Our first episode features Izzy Miller (@isidoremiller), AI Engineer at Hex (@_hex_tech). Hex has been shipping data agents since before most teams were even thinking about them, starting with single-cell text-to-SQL and graduating to a full Notebook agent that can work autonomously for 20 minutes on a complex analysis. Izzy has a lot of perspective on what it actually takes to get agents working well in production, and what breaks along the way. A few takeaways from our conversation: - Keep your eval sets small enough to hold in your head: Izzy runs 30-50 handcrafted "traps" with multiple repetitions, rather than hundreds of variants. If you can't explain why your agent fails each one, your eval set is too big - Day zero performance is almost irrelevant: The more interesting question is how the agent compounds. Izzy is building a 90-day simulation where the warehouse evolves and the agent has to accumulate understanding - You can catch agent errors without seeing the raw outputs: By running an LLM-as-a-judge over production usage and clustering the results, you can surface places where something likely went wrong, without needing to read individual conversations Watch the full episode on: - Youtube: https://t.co/AdkQbV3Pq2 - Apple Podcasts: https://t.co/1MKF7mcYSr - Spotify: https://t.co/DxACw24oob

Media 1
Media 2

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2042279493050740916/media_1.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2042279493050740916/media_1.jpg",
      "type": "photo",
      "filename": "media_1.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2042279493050740916/media_2.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2042279493050740916/media_2.jpg",
      "type": "photo",
      "filename": "media_2.jpg"
    }
  ],
  "processed_at": "2026-04-09T17:21:56.293850",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2042279493050740916",
  "url": "https://x.com/hwchase17/status/2042279493050740916",
  "twitterUrl": "https://twitter.com/hwchase17/status/2042279493050740916",
  "text": "🎙️Introducing Max Agency\n\nMax Agency is a new podcast where we go deep on how the best agents are actually being built: architecture decisions, tradeoffs, evals, and everything in between. Each episode, I sit down with engineering leaders who are doing this work in production.\n\nOur first episode features Izzy Miller (@isidoremiller), AI Engineer at Hex (@_hex_tech). Hex has been shipping data agents since before most teams were even thinking about them, starting with single-cell text-to-SQL and graduating to a full Notebook agent that can work autonomously for 20 minutes on a complex analysis.\n\nIzzy has a lot of perspective on what it actually takes to get agents working well in production, and what breaks along the way.\n\nA few takeaways from our conversation:\n- Keep your eval sets small enough to hold in your head: Izzy runs 30-50 handcrafted \"traps\" with multiple repetitions, rather than hundreds of variants. If you can't explain why your agent fails each one, your eval set is too big\n- Day zero performance is almost irrelevant: The more interesting question is how the agent compounds. Izzy is building a 90-day simulation where the warehouse evolves and the agent has to accumulate understanding\n- You can catch agent errors without seeing the raw outputs: By running an LLM-as-a-judge over production usage and clustering the results, you can surface places where something likely went wrong, without needing to read individual conversations\n\nWatch the full episode on:\n\n- Youtube: https://t.co/AdkQbV3Pq2\n- Apple Podcasts: https://t.co/1MKF7mcYSr\n- Spotify: https://t.co/DxACw24oob",
  "source": "Twitter for iPhone",
  "retweetCount": 12,
  "replyCount": 4,
  "likeCount": 40,
  "quoteCount": 4,
  "viewCount": 1731,
  "createdAt": "Thu Apr 09 16:32:29 +0000 2026",
  "lang": "en",
  "bookmarkCount": 22,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2042279493050740916",
  "displayTextRange": [
    0,
    277
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "hwchase17",
    "url": "https://x.com/hwchase17",
    "twitterUrl": "https://twitter.com/hwchase17",
    "id": "2728439146",
    "name": "Harrison Chase",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1569345624935485442/R67C4wCQ_normal.jpg",
    "coverPicture": "",
    "description": "",
    "location": "",
    "followers": 95635,
    "following": 549,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Tue Jul 29 19:22:50 +0000 2014",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 15233,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 609,
    "statusesCount": 15681,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2040467997022884194"
    ],
    "profile_bio": {
      "description": "@LangChain, previously @robusthq @kensho\n\nMLOps ∪ Generative AI ∪ sports analytics",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [],
          "user_mentions": [
            {
              "id_str": "0",
              "indices": [
                0,
                10
              ],
              "name": "",
              "screen_name": "LangChain"
            },
            {
              "id_str": "0",
              "indices": [
                23,
                32
              ],
              "name": "",
              "screen_name": "robusthq"
            },
            {
              "id_str": "0",
              "indices": [
                33,
                40
              ],
              "name": "",
              "screen_name": "kensho"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "additional_media_info": {
          "monetizable": true
        },
        "display_url": "pic.twitter.com/yqBsBcGQOR",
        "expanded_url": "https://twitter.com/hwchase17/status/2042279493050740916/video/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "id_str": "2042279071934279680",
        "indices": [
          278,
          301
        ],
        "media_key": "13_2042279071934279680",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwABAoAARxXof7W23AAAAA=",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAAECgABHFeh/tbbcAAAAA==",
            "media_key": "13_2042279071934279680"
          }
        },
        "media_url_https": "https://pbs.twimg.com/amplify_video_thumb/2042279071934279680/img/a2lNWW1wLgdbeQDn.jpg",
        "original_info": {
          "focus_rects": [],
          "height": 2160,
          "width": 3840
        },
        "sizes": {
          "large": {
            "h": 1152,
            "w": 2048
          }
        },
        "type": "video",
        "url": "https://t.co/yqBsBcGQOR",
        "video_info": {
          "aspect_ratio": [
            16,
            9
          ],
          "duration_millis": 99560,
          "variants": [
            {
              "content_type": "application/x-mpegURL",
              "url": "https://video.twimg.com/amplify_video/2042279071934279680/pl/-j53CjHer19t2XBH.m3u8?tag=21"
            },
            {
              "bitrate": 256000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2042279071934279680/vid/avc1/480x270/s3v8mnpl_qS1_X6O.mp4?tag=21"
            },
            {
              "bitrate": 832000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2042279071934279680/vid/avc1/640x360/hPPasqJvPMTD_Xek.mp4?tag=21"
            },
            {
              "bitrate": 2176000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2042279071934279680/vid/avc1/1280x720/RW0roB-Kc_IboQ4P.mp4?tag=21"
            },
            {
              "bitrate": 10368000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2042279071934279680/vid/avc1/1920x1080/qkYSOCUsvx0G34n0.mp4?tag=21"
            },
            {
              "bitrate": 25128000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2042279071934279680/vid/avc1/3840x2160/5OLIekq0_j75fpnp.mp4?tag=21"
            }
          ]
        }
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "timestamps": [],
    "urls": [
      {
        "display_url": "youtube.com/watch?v=Xyh1Eq…",
        "expanded_url": "https://www.youtube.com/watch?v=Xyh1EqcjGME",
        "indices": [
          1503,
          1526
        ],
        "url": "https://t.co/AdkQbV3Pq2"
      },
      {
        "display_url": "podcasts.apple.com/us/podcast/how…",
        "expanded_url": "https://podcasts.apple.com/us/podcast/how-hex-builds-ai-agents-making-agents-reason-like/id1891551672?i=1000760489140",
        "indices": [
          1545,
          1568
        ],
        "url": "https://t.co/1MKF7mcYSr"
      },
      {
        "display_url": "open.spotify.com/episode/1BJlg3…",
        "expanded_url": "https://open.spotify.com/episode/1BJlg3SOJrjnaPXFHTNuux?si=bffc89cb4f774617",
        "indices": [
          1580,
          1603
        ],
        "url": "https://t.co/DxACw24oob"
      }
    ],
    "user_mentions": [
      {
        "id_str": "1044353516095979520",
        "indices": [
          319,
          333
        ],
        "name": "Izzy",
        "screen_name": "isidoremiller"
      },
      {
        "id_str": "1187089160345014272",
        "indices": [
          356,
          366
        ],
        "name": "Hex",
        "screen_name": "_hex_tech"
      }
    ]
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}