{
  "apiVersion": "v1",
  "person": {
    "id": "lee-robinson",
    "name": "Lee Robinson",
    "role": "Machine Learning Engineer, Model Behavior, Cursor",
    "level": "practitioner",
    "levelMeaning": "Someone doing the work at the organization.",
    "organization": "Cursor (Anysphere)",
    "organizationId": "cursor",
    "passages": 12,
    "reported": 0,
    "url": "https://the-ai-record.com/people/lee-robinson",
    "json": "https://the-ai-record.com/api/v1/people/lee-robinson.json",
    "background": "Works on model research and behavior at Cursor and was introduced on stage as a machine learning engineer on model behavior; the talk's YouTube description says he was previously at Vercel and the title lists Cursor and SpaceXAI. By October 2026 he was working on Grok Bot at SpaceX: in the CS146S talk he says 'at Cursor and now at SpaceX', and his own site leerob.com says he works on ML at SpaceX, formerly at Cursor.",
    "whyCredible": "Works on the team that trains the Cursor models he describes, so he is close to the training process; Cursor sells AI coding tools and its own models, and the talk is promotional of its approach.",
    "linkedin": null,
    "x": null,
    "profileSource": "https://www.youtube.com/watch?v=q4Tr-DknG2M"
  },
  "passages": [
    {
      "id": "spacex-bots-joining-meetings-take-notes-2026",
      "text": "It's increasingly common now at SpaceX where we see bots joining meetings for employees … but they want their bot to go and just take notes and come back to them, which has gotten funny.",
      "context": "Robinson is listing the events that wake an always-on agent, such as a Slack message or a meeting. The cut skips 'where they could make it to the meeting', which both passes transcribed the same way but which reads as a mis-hearing.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "11:26",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "adoption-and-usage",
        "ai-across-the-company"
      ],
      "functions": [
        "company-wide"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-bots-joining-meetings-take-notes-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-bots-joining-meetings-take-notes-2026.json"
    },
    {
      "id": "spacex-delete-the-product-models-worst-theyll-ever-be-2026",
      "text": "One of the principles that we had at Cursor and now at SpaceX is to delete the product. And what we mean by this is you kind of have to internalize and assume that the models today are the worst they'll ever be, which means that in six months you might need to completely rebuild the UI.",
      "context": "Robinson is describing the product principle his teams followed at Cursor and now at SpaceX while building harnesses for always-on agents.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "37:45",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "product-and-design"
      ],
      "functions": [
        "product",
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-delete-the-product-models-worst-theyll-ever-be-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-delete-the-product-models-worst-theyll-ever-be-2026.json"
    },
    {
      "id": "spacex-grok-bot-agent-computer-firecracker-vm-2026",
      "text": "So I talked about how the agent has this computer, this Linux box. This is running a firecracker virtual machine, and the reason this is really helpful is because it's very easy then to essentially snapshot the state of the machine and shut it down and start it back up.",
      "context": "Robinson is describing the infrastructure behind Grok Bot, the always-on agent his team at SpaceX builds, and the computer each agent gets in the cloud.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "12:24",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "where-agents-run"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-grok-bot-agent-computer-firecracker-vm-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-grok-bot-agent-computer-firecracker-vm-2026.json"
    },
    {
      "id": "spacex-grok-bot-compact-while-cache-warm-2026",
      "text": "One trick that we do, which I think is really helpful, is while you're still within the warm cache period, you should probably do the compaction right now, because you have this long conversation … going to be much cheaper. So just go ahead, do the compaction, and now when the user comes back at 30 minutes or whatever, they're starting from this summary, and the summary now is going to be much cheaper for them to continue adding messages on.",
      "context": "Robinson is describing how his team keeps the cost of long agent conversations down by summarizing the context before the model provider's prompt cache expires.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "32:20",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "ai-cost-and-spend",
        "controlling-ai-spend"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-grok-bot-compact-while-cache-warm-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-grok-bot-compact-while-cache-warm-2026.json"
    },
    {
      "id": "spacex-grok-bot-memory-core-facts-logs-scratchpad-2026",
      "text": "We tried to model some of that here where the harness has, of course, these core facts. It also has these dated logs of kind of your past conversation history centrally. And then also these kind of scratch pad notes … little things that you've been working on right now that are designed to be ephemeral, and they will go away, but they might be relevant just for the next 30 minutes or an hour.",
      "context": "Robinson has just compared the agent's memory to how a brain keeps relevant facts in mind, and is describing the three memory layers in the Grok Bot harness.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "33:51",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "internal-tools",
        "context-for-agents"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-grok-bot-memory-core-facts-logs-scratchpad-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-grok-bot-memory-core-facts-logs-scratchpad-2026.json"
    },
    {
      "id": "spacex-grok-bot-one-tool-client-server-2026",
      "text": "… there's only one tool that communicates between the client and the server. And it's just send user. And the great thing about this is that it's actually very architecturally simple where the client is just texting the server in some ways.",
      "context": "Robinson is describing how the Grok Bot harness differs from other tools on the market. The quote starts mid-sentence, after a lead-in the two transcription passes heard differently.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "18:21",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "where-agents-run",
        "internal-tools"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-grok-bot-one-tool-client-server-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-grok-bot-one-tool-client-server-2026.json"
    },
    {
      "id": "spacex-grok-bot-temporal-durable-workflows-2026",
      "text": "So we use an open source tool called temporal, which is interesting to look into if you're curious, so that we don't have to rebuild a lot of this very complicated stuff from scratch.",
      "context": "Robinson has just explained that an agent task that fails midway should resume at the failed step, not replay earlier steps, and says Grok Bot uses Temporal for these durable workflows.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Works on Grok Bot at SpaceX (formerly Cursor)",
      "voice": "person",
      "organization": "SpaceX (SpaceXAI)",
      "organizationId": "spacex",
      "organizationUrl": "https://the-ai-record.com/organizations/spacex",
      "aiVendor": true,
      "work": {
        "title": "Stanford CS146S: How always-on agents work",
        "type": "talk",
        "date": "2026-10-11",
        "publisher": "leerob (YouTube)",
        "locator": "17:08",
        "url": "https://www.youtube.com/watch?v=L1OBRtfPS34"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-07"
      },
      "topics": [
        "internal-tools",
        "where-agents-run"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/spacex-grok-bot-temporal-durable-workflows-2026",
      "json": "https://the-ai-record.com/api/v1/passages/spacex-grok-bot-temporal-durable-workflows-2026.json"
    },
    {
      "id": "cursor-bench-private-eval-own-codebase-2026",
      "text": "And that's why we have Cursor Bench. We have this private eval set that is mostly made up of things that happen in our code base, which is held out from the … So we ensure that the models aren't trained on it. And it's based on those real world engineering tasks.",
      "context": "Robinson has just said public benchmarks are an imperfect test of how models are used in practice, and is describing Cursor's private evaluation set for its own models.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Machine Learning Engineer, Model Behavior, Cursor",
      "voice": "person",
      "organization": "Cursor (Anysphere)",
      "organizationId": "cursor",
      "organizationUrl": "https://the-ai-record.com/organizations/cursor",
      "aiVendor": true,
      "work": {
        "title": "Recursive Model Improvement",
        "type": "talk",
        "date": "2026-07-15",
        "publisher": "AI Engineer (YouTube)",
        "locator": "8:12",
        "url": "https://www.youtube.com/watch?v=q4Tr-DknG2M"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-03"
      },
      "topics": [
        "measuring-impact",
        "choosing-models",
        "ai-evals"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/cursor-bench-private-eval-own-codebase-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cursor-bench-private-eval-own-codebase-2026.json"
    },
    {
      "id": "cursor-composer-2-5-most-popular-model-2026",
      "text": "So we put out Composer 2.5 in May. And it's now the most popular model in Cursor, which is exciting. And we scaled up training here quite a bit by generating more RL environments, trying out some new methods for learning. And also just making more ambitious problems for the models to solve. And the results have been pretty promising so far.",
      "context": "Robinson is describing Cursor's own model, Composer, which Cursor trains in-house.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Machine Learning Engineer, Model Behavior, Cursor",
      "voice": "person",
      "organization": "Cursor (Anysphere)",
      "organizationId": "cursor",
      "organizationUrl": "https://the-ai-record.com/organizations/cursor",
      "aiVendor": true,
      "work": {
        "title": "Recursive Model Improvement",
        "type": "talk",
        "date": "2026-07-15",
        "publisher": "AI Engineer (YouTube)",
        "locator": "2:41",
        "url": "https://www.youtube.com/watch?v=q4Tr-DknG2M"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-03"
      },
      "topics": [
        "adoption-and-usage"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/cursor-composer-2-5-most-popular-model-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cursor-composer-2-5-most-popular-model-2026.json"
    },
    {
      "id": "cursor-model-learning-to-train-next-model-2026",
      "text": "The last bit here is that the model is learning to train the next model. And it's a little hard to wrap your brain around. The way I like to think about it is every time you release a new version of this intelligence, then you can create or distill these derivative versions that you use to speed up other parts of the training process, …",
      "context": "Robinson is closing a talk on how Cursor trains its models and uses earlier models to speed up later training.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Machine Learning Engineer, Model Behavior, Cursor",
      "voice": "person",
      "organization": "Cursor (Anysphere)",
      "organizationId": "cursor",
      "organizationUrl": "https://the-ai-record.com/organizations/cursor",
      "aiVendor": true,
      "work": {
        "title": "Recursive Model Improvement",
        "type": "talk",
        "date": "2026-07-15",
        "publisher": "AI Engineer (YouTube)",
        "locator": "18:30",
        "url": "https://www.youtube.com/watch?v=q4Tr-DknG2M"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-03"
      },
      "topics": [
        "internal-tools",
        "ai-across-the-company"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/cursor-model-learning-to-train-next-model-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cursor-model-learning-to-train-next-model-2026.json"
    },
    {
      "id": "cursor-researchers-run-experiments-from-slack-2026",
      "text": "Just to kind of put an example to this, we've created these tools and these systems where researchers can run experiments directly from Slack. We … avoid this state of being bottlenecked on humans launching, and reviewing, and babysitting runs. And we actually have an entire team just working on automating every part of the research work that isn't freeing up the researchers' time to work on their most ambitious ideas. So every person on the ML team gets access to this fleet of agents they can basically train models directly from Slack.",
      "context": "Robinson is describing how Cursor's machine learning team uses agents in its own model-training work.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Machine Learning Engineer, Model Behavior, Cursor",
      "voice": "person",
      "organization": "Cursor (Anysphere)",
      "organizationId": "cursor",
      "organizationUrl": "https://the-ai-record.com/organizations/cursor",
      "aiVendor": true,
      "work": {
        "title": "Recursive Model Improvement",
        "type": "talk",
        "date": "2026-07-15",
        "publisher": "AI Engineer (YouTube)",
        "locator": "17:14",
        "url": "https://www.youtube.com/watch?v=q4Tr-DknG2M"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-03"
      },
      "topics": [
        "internal-tools",
        "autonomy-and-oversight"
      ],
      "functions": [
        "engineering"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/cursor-researchers-run-experiments-from-slack-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cursor-researchers-run-experiments-from-slack-2026.json"
    },
    {
      "id": "cursor-vast-majority-revenue-agent-usage-2026",
      "text": "And in reality, the vast majority of our revenue today comes from agent usage. And that means that all of the data inside of Cursor is also coming from agent usage, and we can use that to train better models.",
      "context": "Robinson is explaining where the data for training Cursor's models comes from. The talk gives no revenue figure.",
      "speaker": "Lee Robinson",
      "speakerId": "lee-robinson",
      "speakerUrl": "https://the-ai-record.com/people/lee-robinson",
      "speakerLevel": "practitioner",
      "role": "Machine Learning Engineer, Model Behavior, Cursor",
      "voice": "person",
      "organization": "Cursor (Anysphere)",
      "organizationId": "cursor",
      "organizationUrl": "https://the-ai-record.com/organizations/cursor",
      "aiVendor": true,
      "work": {
        "title": "Recursive Model Improvement",
        "type": "talk",
        "date": "2026-07-15",
        "publisher": "AI Engineer (YouTube)",
        "locator": "4:41",
        "url": "https://www.youtube.com/watch?v=q4Tr-DknG2M"
      },
      "grade": "B",
      "gradeMeaning": "The organization's own post, paper or talk.",
      "verification": {
        "status": "verified",
        "meaning": "An independent second reader re-fetched the source and confirmed the wording.",
        "reviewedOn": "2026-11-03"
      },
      "topics": [
        "adoption-and-usage",
        "coding-agents"
      ],
      "functions": [
        "product"
      ],
      "figures": [],
      "url": "https://the-ai-record.com/passages/cursor-vast-majority-revenue-agent-usage-2026",
      "json": "https://the-ai-record.com/api/v1/passages/cursor-vast-majority-revenue-agent-usage-2026.json"
    }
  ],
  "reportedPassages": [],
  "notice": [
    "Every passage is reproduced word for word from the cited source.",
    "Check the verification status: \"reported\" means not yet confirmed by an independent second reader.",
    "Organizations flagged aiVendor sell AI tools or services and have a reason to publish flattering numbers.",
    "An empty result means the archive holds nothing on it, not that nobody has said anything."
  ]
}