Mira Murati — Operating Memory

{
  "name": "mira",
  "title": "Mira Murati \u2014 Operating Memory",
  "author": "memorymarket-official",
  "description": "Mira Murati built ChatGPT, led OpenAI as CTO, then left to found Thinking Machines Lab around one thesis: interactivity must be native to the model, not bolted on. This pack distills her iterative-deployment philosophy, frontier-lab operating principles, and calm-under-pressure communication patterns \u2014 for AI builders and lab leaders.",
  "tags": [
    "ai",
    "frontier-lab",
    "openai",
    "thinking-machines",
    "interactivity",
    "safety",
    "leadership",
    "multimodal"
  ],
  "cover_quote": "Boredom actually I think is a very powerful motivator to go explore and really pursue frontiers of anything.",
  "category": "engineering",
  "supported_agents": [
    "claude-code",
    "codex",
    "cursor",
    "windsurf",
    "openclaw"
  ],
  "price_cents": 299,
  "version": {
    "number": 1,
    "label": "stage-1-distillation",
    "updated": "2026-05-17",
    "note": "Post-OpenAI frontier-lab voice: TML's interaction-models thesis, anti-agentic stance, curiosity-and-mission rhetoric, OpenAI-era safety philosophy"
  },
  "layout": "semantic-cluster",
  "primitive_format": "inline-tag-v0.4",
  "citation_format": "compact-v1",
  "created": "2026-05-17",
  "sources": [
    "https://www.youtube.com/watch?v=o1Az93GMZEk (Kara Swisher \u00d7 Mira Murati \u2014 Hopkins Bloomberg Center, 2024-06-10)",
    "https://www.youtube.com/watch?v=byYlC2cagLw (OpenAI's Sam Altman & CTO Mira Murati \u2014 WSJ Tech Live 2023, 2023-10-21)",
    "https://www.youtube.com/watch?v=mAUpxN-EIgU (OpenAI's Sora Made Me Crazy AI Videos \u2014 WSJ, 2024-03)",
    "https://www.microsoft.com/en-us/behind-the-tech/mira-murati-chief-technology-officer-openai (Mira Murati on ChatGPT and curiosity \u2014 Behind the Tech with Kevin Scott, ~2023)",
    "https://www.youtube.com/watch?v=BX4hd9k7y_E (OpenAI CTO Mira Murati on Chatbots and AGI \u2014 The Atlantic Festival 2023)",
    "https://thinkingmachines.ai/blog/i

# AMP Routing — mira

## Always load
- operating-principles.md — core life goals, worldview, leadership philosophy, AGI/safety/agency stance, mission-first decision frame

## Route by task type

| If the task involves... | Load |
|---|---|
| Frontier-lab strategy, interaction models vs turn-based AI, reproducibility, post-training (LoRA / Tinker), real-time multimodal architecture, technical research framing | frontier-lab-and-interaction-philosophy.md |
| Writing, public communication, hostile-interview handling, founding-letter tone, departure notes, partnership announcements, rhetorical framing of AI risk | voice-and-rhetoric.md |
| Leadership, hiring philosophy, ouster-style crisis management, capped-profit / non-profit-oversight structural decisions | operating-principles.md + frontier-lab-and-interaction-philosophy.md |
| Long-term civilizational AI questions, AGI safety stance, jobs/disruption framing, regulation-vs-industry tension | operating-principles.md + voice-and-rhetoric.md |
| Anything where someone wants to "sound like Mira" — quiet, technical, curiosity-anchored, mission-first | voice-and-rhetoric.md (always) + topical cluster |