{
  "docId": "019dd923-605c-759f-b6b2-20dfed71ff5d",
  "docSlug": "bi-f0c242eb2eadbf4a",
  "documentTitle": "The pitch deck machine-learning firm Galileo used to raise $18 million",
  "authorId": "Pitchdecks",
  "authorName": "Galileo",
  "documentKindSlug": "pitchdeck",
  "documentKindLabel": "Pitch deck",
  "sourceTypeSlug": "strategy_consulting",
  "sourceTypeLabel": "Strategy consulting",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.784,
  "pageNumber": 5,
  "pageCount": 12,
  "prevPage": 4,
  "nextPage": 6,
  "slideType": "problem_statement",
  "function": "frame_problem",
  "density": "dense",
  "nDataPoints": 0,
  "notes": "The slide uses a process-flow diagram to illustrate where specific user frustrations occur in the ML pipeline.",
  "elementsJson": [
    "headline_text",
    "process_diagram",
    "icon_grid"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-605c-759f-b6b2-20dfed71ff5d/5",
  "deckHref": "/decks/019dd923-605c-759f-b6b2-20dfed71ff5d",
  "deckJsonHref": "/decks/019dd923-605c-759f-b6b2-20dfed71ff5d.json",
  "deckAnchorHref": "/decks/019dd923-605c-759f-b6b2-20dfed71ff5d#slide-5",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "1: How do I inspect my unlabeled data to find errors fast? 2: It takes weeks of 'data detective work' to find the data that my model did poorly on. 3: I am blind to how my model is reacting to production data.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-cb18-702b-89cc-093c19c5c6ab",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.7,
        "x": 0.15,
        "y": 0.55
      },
      "kind": "diagram",
      "text": "Flow from Training Data to Well Trained Model to Predictions",
      "attrs": null,
      "subkind": "process",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "3182e488-5fbd-4447-905d-014ae7aba627",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.2,
        "x": 0.55,
        "y": 0.78
      },
      "kind": "paragraph",
      "text": "2: It takes weeks of 'data detective work' to find the data that my model did poorly on",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "42da7f95-d030-4b67-afc4-554d1d310039",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.2,
        "x": 0.25,
        "y": 0.78
      },
      "kind": "paragraph",
      "text": "1: How do I inspect my unlabeled data to find errors fast?",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "9e0398f9-900a-49f4-a35d-3fccae70e55c",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.34,
        "x": 0.33,
        "y": 0.35
      },
      "kind": "paragraph",
      "text": "3: I am blind to how my model is reacting to production data. The world changes, and I need to know how my model needs to adapt",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d5d9aedf-d8aa-47d4-8d67-81cfc923deea",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.06,
        "w": 0.76,
        "x": 0.12,
        "y": 0.08
      },
      "kind": "title",
      "text": "3 Critical Data Challenges across the ML workflow today",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "9dd9c541-31b8-4763-948a-6bf188a2d350",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 5,
      "from": 4,
      "beatId": "12bd29df-709a-4d3e-80f7-8cbc8d87f1aa",
      "arcName": "Problem-Agitate-Solution",
      "arcSlug": "problem-agitate-solution",
      "beatName": "Agitation",
      "beatSlug": "problem-agitate-solution-agitate-make-it-worse",
      "evidence": "Slides 4-5 agitate the problem",
      "position": 1,
      "confidence": 0.9,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 5,
      "from": 4,
      "name": "Cost Of Inaction",
      "slug": "27-cost-of-inaction",
      "bestFor": "Urgent budget requests, compliance, risk mitigation",
      "matchId": "5c2dbacd-8ce8-4cf7-aab1-9e6354ad3fbc",
      "evidence": "Slides 4-5 imply the costs",
      "position": 0,
      "objective": "What are the costs of poor data quality?",
      "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point",
      "confidence": 0.7,
      "description": "Quantify what happens if the audience does nothing"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}