🐦 Twitter Post Details

Viewing enriched Twitter post

@jerryjliu0

This is a great tutorial (credits @itsclelia + @lancedb) on how to build a practical retrieval pipeline that integrates directly with your agent harness. 1. Ingest a massive pile of docs with liteparse. 2. Store data in a vector db (despite my memes to the contrary, you will need some database for larger scale retrieval). 3. Pair with image screenshotting tools that allow the agent to "dive deeper" into data. When you pair this with the Claude Agent SDK / Claude Code, the agent will do some initial retrieval pass to pull the relevant doc, and then use screenshotting/VLM-enabled capabilities to do deeper analysis. Blog: https://t.co/mbdy4I5oAh

Media 1
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2041665979261108418/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2041665979261108418/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-04-08T14:31:20.950272",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2041665979261108418",
  "url": "https://x.com/jerryjliu0/status/2041665979261108418",
  "twitterUrl": "https://twitter.com/jerryjliu0/status/2041665979261108418",
  "text": "This is a great tutorial (credits @itsclelia + @lancedb) on how to build a practical retrieval pipeline that integrates directly with your agent harness.\n\n1. Ingest a massive pile of docs with liteparse.\n2. Store data in a vector db (despite my memes to the contrary, you will need some database for larger scale retrieval).\n3. Pair with image screenshotting tools that allow the agent to \"dive deeper\" into data. \n\nWhen you pair this with the Claude Agent SDK / Claude Code, the agent will do some initial retrieval pass to pull the relevant doc, and then use screenshotting/VLM-enabled capabilities to do deeper analysis.\n\nBlog: https://t.co/mbdy4I5oAh",
  "source": "Twitter for iPhone",
  "retweetCount": 15,
  "replyCount": 1,
  "likeCount": 101,
  "quoteCount": 0,
  "viewCount": 11011,
  "createdAt": "Tue Apr 07 23:54:36 +0000 2026",
  "lang": "en",
  "bookmarkCount": 106,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2041665979261108418",
  "displayTextRange": [
    0,
    276
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "jerryjliu0",
    "url": "https://x.com/jerryjliu0",
    "twitterUrl": "https://twitter.com/jerryjliu0",
    "id": "369777416",
    "name": "Jerry Liu",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1283610285031460864/1Q4zYhtb_normal.jpg",
    "coverPicture": "",
    "description": "",
    "location": "",
    "followers": 72346,
    "following": 1468,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Wed Sep 07 22:54:31 +0000 2011",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 8613,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 1473,
    "statusesCount": 6822,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [],
    "profile_bio": {
      "description": "Parsing the world's hardest PDFs @llama_index. cofounder/CEO\n\nCareers: https://t.co/EUnMNmbCtx\nEnterprise: https://t.co/Ht5jwxSrQB",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [
            {
              "display_url": "llamaindex.ai/careers",
              "expanded_url": "https://www.llamaindex.ai/careers",
              "indices": [
                71,
                94
              ],
              "url": "https://t.co/EUnMNmbCtx"
            },
            {
              "display_url": "llamaindex.ai/contact",
              "expanded_url": "https://www.llamaindex.ai/contact",
              "indices": [
                107,
                130
              ],
              "url": "https://t.co/Ht5jwxSrQB"
            }
          ],
          "user_mentions": [
            {
              "id_str": "0",
              "indices": [
                33,
                45
              ],
              "name": "",
              "screen_name": "llama_index"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "llamaindex.ai",
              "expanded_url": "https://www.llamaindex.ai/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/YiIfjVlzb6"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/U9KgJBiMji",
        "expanded_url": "https://twitter.com/jerryjliu0/status/2041665979261108418/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2041665166832758784",
        "indices": [
          277,
          300
        ],
        "media_key": "3_2041665166832758784",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxVc6brWvAACgACHFV0ZBPbwMIAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHFVzputa8AAKAAIcVXRkE9vAwgAA",
            "media_key": "3_2041665166832758784"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HFVzputa8AAqyDG.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 1134,
              "w": 2025,
              "x": 0,
              "y": 0
            },
            {
              "h": 1134,
              "w": 1134,
              "x": 0,
              "y": 0
            },
            {
              "h": 1134,
              "w": 995,
              "x": 0,
              "y": 0
            },
            {
              "h": 1134,
              "w": 567,
              "x": 106,
              "y": 0
            },
            {
              "h": 1134,
              "w": 3118,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1134,
          "width": 3118
        },
        "sizes": {
          "large": {
            "h": 745,
            "w": 2048
          }
        },
        "type": "photo",
        "url": "https://t.co/U9KgJBiMji"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "lancedb.com/blog/smart-par…",
        "expanded_url": "https://www.lancedb.com/blog/smart-parsing-meets-sharp-retrieval-combining-liteparse-and-lancedb",
        "indices": [
          631,
          654
        ],
        "url": "https://t.co/mbdy4I5oAh"
      }
    ],
    "user_mentions": [
      {
        "id_str": "1925178481534509056",
        "indices": [
          34,
          44
        ],
        "name": "Clelia Bertelli (🦙/acc)",
        "screen_name": "itsclelia"
      },
      {
        "id_str": "1648910047270010880",
        "indices": [
          47,
          55
        ],
        "name": "LanceDB",
        "screen_name": "lancedb"
      }
    ]
  },
  "quoted_tweet": {
    "type": "tweet",
    "id": "2041638826091946450",
    "url": "https://x.com/itsclelia/status/2041638826091946450",
    "twitterUrl": "https://twitter.com/itsclelia/status/2041638826091946450",
    "text": "How can you improve your agentic search pipeline?\nI just wrote a blog post with @tech_optimist from @lancedb to answer exactly that.\nTLDR:\n- Parse files and take page-level screenshots with LiteParse, the parser we just open sourced at @llama_index\n- Chunk and embed text, and store everything (text, image bytes, vector data) in a local LanceDB instance\n- Expose text and image retrieval tools to a Claude agent, and let it reason on both data types\n\nWith our eval dataset, the agent got near-perfect scores on most complex QA tasks, showing how a strong parsing foundation and multimodal retrieval can really improve your search🚀\nRead the full breakdown here: https://t.co/5GRRy065TI",
    "source": "Twitter for iPhone",
    "retweetCount": 4,
    "replyCount": 0,
    "likeCount": 20,
    "quoteCount": 1,
    "viewCount": 13746,
    "createdAt": "Tue Apr 07 22:06:42 +0000 2026",
    "lang": "en",
    "bookmarkCount": 28,
    "isReply": false,
    "inReplyToId": null,
    "conversationId": "2041638826091946450",
    "displayTextRange": [
      0,
      276
    ],
    "inReplyToUserId": null,
    "inReplyToUsername": null,
    "author": {
      "type": "user",
      "userName": "itsclelia",
      "url": "https://x.com/itsclelia",
      "twitterUrl": "https://twitter.com/itsclelia",
      "id": "1925178481534509056",
      "name": "Clelia Bertelli (🦙/acc)",
      "isVerified": false,
      "isBlueVerified": false,
      "verifiedType": null,
      "profilePicture": "https://pbs.twimg.com/profile_images/2010129136501555200/gaI8nXYh_normal.jpg",
      "coverPicture": "https://pbs.twimg.com/profile_banners/1925178481534509056/1773305599",
      "description": "",
      "location": "",
      "followers": 793,
      "following": 384,
      "status": "",
      "canDm": false,
      "canMediaTag": true,
      "createdAt": "Wed May 21 13:15:29 +0000 2025",
      "entities": {
        "description": {
          "urls": []
        },
        "url": {}
      },
      "fastFollowersCount": 0,
      "favouritesCount": 1388,
      "hasCustomTimelines": true,
      "isTranslator": false,
      "mediaCount": 145,
      "statusesCount": 796,
      "withheldInCountries": [],
      "affiliatesHighlightedLabel": {},
      "possiblySensitive": false,
      "pinnedTweetIds": [
        "2040111180203892928"
      ],
      "profile_bio": {
        "description": "she/her |\nOpen Sourceress at @llama_index 🦙 | \nGirly dev building AI things in python💅 | Cheating on python with Go and Rust🦫🦀",
        "entities": {
          "description": {
            "hashtags": [],
            "symbols": [],
            "urls": [],
            "user_mentions": [
              {
                "id_str": "0",
                "indices": [
                  29,
                  41
                ],
                "name": "",
                "screen_name": "llama_index"
              }
            ]
          },
          "url": {
            "urls": [
              {
                "display_url": "clelia.dev",
                "expanded_url": "https://clelia.dev",
                "indices": [
                  0,
                  23
                ],
                "url": "https://t.co/zhCyKugORT"
              }
            ]
          }
        }
      },
      "isAutomated": false,
      "automatedBy": null
    },
    "extendedEntities": {
      "media": [
        {
          "display_url": "pic.twitter.com/Sq88vO7H4i",
          "expanded_url": "https://twitter.com/itsclelia/status/2041638826091946450/photo/1",
          "ext_media_availability": {
            "status": "Available"
          },
          "features": {
            "large": {
              "faces": [
                {
                  "h": 69,
                  "w": 69,
                  "x": 1238,
                  "y": 610
                }
              ]
            },
            "orig": {
              "faces": [
                {
                  "h": 69,
                  "w": 69,
                  "x": 1238,
                  "y": 610
                }
              ]
            }
          },
          "id_str": "2041638822937817088",
          "indices": [
            277,
            300
          ],
          "media_key": "3_2041638822937817088",
          "media_results": {
            "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxVW7FA2vAACgACHFVbsfzbIdIAAA==",
            "result": {
              "__typename": "ApiMedia",
              "id": "QXBpTWVkaWE6DAABCgABHFVbsUDa8AAKAAIcVVux/Nsh0gAA",
              "media_key": "3_2041638822937817088"
            }
          },
          "media_url_https": "https://pbs.twimg.com/media/HFVbsUDa8AAubT1.jpg",
          "original_info": {
            "focus_rects": [
              {
                "h": 768,
                "w": 1371,
                "x": 0,
                "y": 0
              },
              {
                "h": 768,
                "w": 768,
                "x": 0,
                "y": 0
              },
              {
                "h": 768,
                "w": 674,
                "x": 0,
                "y": 0
              },
              {
                "h": 768,
                "w": 384,
                "x": 116,
                "y": 0
              },
              {
                "h": 768,
                "w": 1376,
                "x": 0,
                "y": 0
              }
            ],
            "height": 768,
            "width": 1376
          },
          "sizes": {
            "large": {
              "h": 768,
              "w": 1376
            }
          },
          "type": "photo",
          "url": "https://t.co/Sq88vO7H4i"
        }
      ]
    },
    "card": null,
    "place": {},
    "entities": {
      "hashtags": [],
      "symbols": [],
      "urls": [
        {
          "display_url": "lancedb.com/blog/smart-par…",
          "expanded_url": "https://www.lancedb.com/blog/smart-parsing-meets-sharp-retrieval-combining-liteparse-and-lancedb",
          "indices": [
            662,
            685
          ],
          "url": "https://t.co/5GRRy065TI"
        }
      ],
      "user_mentions": [
        {
          "id_str": "1928993071",
          "indices": [
            80,
            94
          ],
          "name": "Prashanth Rao",
          "screen_name": "tech_optimist"
        },
        {
          "id_str": "1648910047270010880",
          "indices": [
            100,
            108
          ],
          "name": "LanceDB",
          "screen_name": "lancedb"
        },
        {
          "id_str": "1604278358296055808",
          "indices": [
            236,
            248
          ],
          "name": "LlamaIndex 🦙",
          "screen_name": "llama_index"
        }
      ]
    },
    "quoted_tweet": null,
    "retweeted_tweet": null,
    "isLimitedReply": false,
    "communityInfo": null,
    "article": null
  },
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}