🐦 Twitter Post Details

Viewing enriched Twitter post

@renegadesilicon

This is exactly why evals matter more than prompt or engineering skills alone....because instruction-following is measurable! Without the eval layer, “I wrote a good skill” can become mostly (sometimes, even dangerous) bs. Evals mean you can turn things like a skill's "hard prohibitions" into automated checks that make model output quantifiable and testable; pretty much every flaw listed here in @lex_node 's tweet would qualify. Code-based evals would catch a lot of these deterministically; LLM-as-judge would handle the more semantic ones. So what can you do to learn evals to fix these types of (literal) skill issues? Where do you start? Well, you can take @HamelHusain and @sh_reya 's excellent evals course! But, to give you some appetite for your deep dive, let's drill into some step by step evals basics for beginners:

📊 Media Metadata

{
  "score": 0.42,
  "score_components": {
    "author": 0.09,
    "engagement": 0.0,
    "quality": 0.12,
    "source": 0.135,
    "nlp": 0.05,
    "recency": 0.025
  },
  "scored_at": "2026-10-21T13:03:05.929225",
  "import_source": "api_import",
  "source_tagged_at": "2026-10-21T13:03:05.929297",
  "enriched": true,
  "enriched_at": "2026-10-21T13:03:05.929300"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2112871296405299545",
  "url": "https://x.com/renegadesilicon/status/2112871296405299545",
  "twitterUrl": "https://twitter.com/renegadesilicon/status/2112871296405299545",
  "text": "This is exactly why evals matter more than prompt or engineering skills alone....because instruction-following is measurable! \n\nWithout the eval layer, “I wrote a good skill” can become mostly (sometimes, even dangerous) bs.\n\nEvals mean you can turn things like a skill's \"hard prohibitions\" into automated checks that make model output quantifiable and testable; pretty much every flaw listed here in @lex_node 's tweet would qualify. Code-based evals would catch a lot of these deterministically; LLM-as-judge would handle the more semantic ones.\n\nSo what can you do to learn evals to fix these types of (literal) skill issues? Where do you start? Well, you can take @HamelHusain and @sh_reya 's excellent evals course! \n\nBut, to give you some appetite for your deep dive, let's drill into some step by step evals basics for beginners:",
  "source": "",
  "retweetCount": 1,
  "replyCount": 2,
  "likeCount": 2,
  "quoteCount": 1,
  "viewCount": 542,
  "createdAt": "Wed Oct 21 11:39:06 +0000 2026",
  "lang": "en",
  "bookmarkCount": 6,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2112871296405299545",
  "displayTextRange": [
    0,
    277
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "renegadesilicon",
    "url": "https://x.com/renegadesilicon",
    "twitterUrl": "https://twitter.com/renegadesilicon",
    "id": "1525242040446881792",
    "name": "Nick",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2012421568694194177/IHOAwjVx_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/1525242040446881792/1757230883",
    "description": "",
    "location": "Australia",
    "followers": 2393,
    "following": 677,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Fri May 13 22:30:29 +0000 2022",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 27091,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 894,
    "statusesCount": 12338,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "1905210012797911410"
    ],
    "profile_bio": {
      "description": "AI/ML R&D | Classics | Bronze Age Med & Nr East | Mathematics | Athletics | Warlord without portfolio | Fighting slop with snark",
      "entities": {
        "description": {}
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {},
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": [
      {
        "id_str": "954015928924057601",
        "indices": [
          402,
          411
        ],
        "name": "gabriel shapiro",
        "screen_name": "lex_node"
      },
      {
        "id_str": "825766640",
        "indices": [
          669,
          681
        ],
        "name": "Hamel Husain",
        "screen_name": "HamelHusain"
      },
      {
        "id_str": "2286218053",
        "indices": [
          686,
          694
        ],
        "name": "Shreya Shankar",
        "screen_name": "sh_reya"
      }
    ]
  },
  "quoted_tweet": {
    "type": "tweet",
    "id": "2112751127712104869",
    "url": "https://x.com/lex_node/status/2112751127712104869",
    "twitterUrl": "https://twitter.com/lex_node/status/2112751127712104869",
    "text": "this is my issue with AI x law. the last thing I want as a lawyer is anything non-deterministic in my documents lol. it doesn't obey the skills even if you wrote the skills to try to make it behave \n\ngive me a rigid docgen template and an AI that can't do anything but input variables into the docgen, over generalized legal AI, any day of the week . . .",
    "source": "",
    "retweetCount": 3,
    "replyCount": 18,
    "likeCount": 50,
    "quoteCount": 2,
    "viewCount": 4167,
    "createdAt": "Wed Oct 21 03:41:36 +0000 2026",
    "lang": "en",
    "bookmarkCount": 11,
    "isReply": false,
    "inReplyToId": null,
    "conversationId": "2112751127712104869",
    "displayTextRange": [
      0,
      274
    ],
    "inReplyToUserId": null,
    "inReplyToUsername": null,
    "author": {
      "type": "user",
      "userName": "lex_node",
      "url": "https://x.com/lex_node",
      "twitterUrl": "https://twitter.com/lex_node",
      "id": "954015928924057601",
      "name": "gabriel shapiro",
      "isVerified": false,
      "isBlueVerified": true,
      "verifiedType": null,
      "profilePicture": "https://pbs.twimg.com/profile_images/2058233189882032128/pRSMN_6C_normal.jpg",
      "coverPicture": "https://pbs.twimg.com/profile_banners/954015928924057601/1779556009",
      "description": "",
      "location": "Cybertron",
      "followers": 78391,
      "following": 8255,
      "status": "",
      "canDm": true,
      "canMediaTag": false,
      "createdAt": "Thu Jan 18 15:41:37 +0000 2018",
      "entities": {
        "description": {
          "urls": []
        },
        "url": {}
      },
      "fastFollowersCount": 0,
      "favouritesCount": 155090,
      "hasCustomTimelines": true,
      "isTranslator": false,
      "mediaCount": 6300,
      "statusesCount": 79850,
      "withheldInCountries": [],
      "affiliatesHighlightedLabel": {
        "label": {
          "badge": {
            "url": "https://pbs.twimg.com/profile_images/1878143326542614528/zi799uOu_bigger.jpg"
          },
          "description": "MetaLeX",
          "url": {
            "url": "https://twitter.com/MetaLeX_Labs",
            "url_type": "DeepLink"
          },
          "user_label_display_type": "Badge",
          "user_label_type": "BusinessLabel"
        }
      },
      "possiblySensitive": false,
      "pinnedTweetIds": [
        "2094405696624148821"
      ],
      "profile_bio": {
        "description": "attorney practicing cybernetic law @MetaLeX_Labs\n\npast: BigLaw Buyside Tech M&A (Weil, Hogan Lovells), @Delphi_Labs GC, @BrownUniversity \n\n$ETH / $GOR barbell",
        "entities": {
          "description": {
            "symbols": [
              {
                "indices": [
                  139,
                  143
                ],
                "text": "ETH"
              },
              {
                "indices": [
                  146,
                  150
                ],
                "text": "GOR"
              }
            ],
            "user_mentions": [
              {
                "id_str": "",
                "indices": [
                  35,
                  48
                ],
                "name": "",
                "screen_name": "MetaLeX_Labs"
              },
              {
                "id_str": "",
                "indices": [
                  103,
                  115
                ],
                "name": "",
                "screen_name": "Delphi_Labs"
              },
              {
                "id_str": "",
                "indices": [
                  120,
                  136
                ],
                "name": "",
                "screen_name": "BrownUniversity"
              }
            ]
          },
          "url": {
            "urls": [
              {
                "display_url": "cyberraise.metalex.tech",
                "expanded_url": "https://cyberraise.metalex.tech/",
                "indices": [
                  0,
                  23
                ],
                "url": "https://t.co/djfvP4QCKA"
              }
            ]
          }
        }
      },
      "isAutomated": false,
      "automatedBy": null
    },
    "extendedEntities": {
      "media": [
        {
          "allow_download_status": {
            "allow_download": true
          },
          "display_url": "pic.x.com/FjyjZVhtDV",
          "expanded_url": "https://x.com/lex_node/status/2112751127712104869/photo/1",
          "ext_master_playlist_only": [],
          "ext_media_availability": {
            "status": "Available"
          },
          "ext_playlists": [],
          "features": {
            "large": {
              "faces": []
            },
            "orig": {
              "faces": []
            }
          },
          "id_str": "2112750749457281024",
          "indices": [
            275,
            298
          ],
          "media_key": "3_2112750749457281024",
          "media_results": {
            "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAAR1R/5zgV3AACgACHVH/9PIWAaUAAA==",
            "result": {
              "__typename": "ApiMedia",
              "id": "QXBpTWVkaWE6DAABCgABHVH/nOBXcAAKAAIdUf/08hYBpQAA",
              "media_key": "3_2112750749457281024"
            }
          },
          "media_url_https": "https://pbs.twimg.com/media/HVH_nOBXcAA3Tjb.png",
          "original_info": {
            "focus_rects": [
              {
                "h": 344,
                "w": 614,
                "x": 0,
                "y": 0
              },
              {
                "h": 344,
                "w": 344,
                "x": 0,
                "y": 0
              },
              {
                "h": 344,
                "w": 302,
                "x": 6,
                "y": 0
              },
              {
                "h": 344,
                "w": 172,
                "x": 71,
                "y": 0
              },
              {
                "h": 344,
                "w": 900,
                "x": 0,
                "y": 0
              }
            ],
            "height": 344,
            "width": 900
          },
          "sizes": {
            "large": {
              "h": 344,
              "w": 900
            }
          },
          "type": "photo",
          "url": "https://t.co/FjyjZVhtDV"
        }
      ]
    },
    "card": null,
    "place": {},
    "entities": {
      "hashtags": [],
      "symbols": [],
      "urls": [],
      "user_mentions": []
    },
    "quoted_tweet": null,
    "retweeted_tweet": null,
    "isLimitedReply": false,
    "communityInfo": null,
    "article": null
  },
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}