{
  "id": 542349,
  "title": "uv talk at Jane Street",
  "url": "/competitions/jane-street-real-time-market-data-forecasting/discussion/542349",
  "author_name": "",
  "post_date": "2024-10-24T10:40:09.869918400Z",
  "votes": 7,
  "comment_count": 4,
  "views": 0,
  "content": "<p>Jane Street recently uploaded this presentation by Charlie Marsh (astral.sh) about how and why he built the popular Python package manager <code>uv</code>. It's great!</p>\n<p><a href=\"https://www.youtube.com/watch?v=gSKTfG1GXYQ\" target=\"_blank\">https://www.youtube.com/watch?v=gSKTfG1GXYQ</a></p>",
  "messages": [
    {
      "id": "3026912",
      "postDate": "10/24/2024 10:40:09",
      "content": "<p>Jane Street recently uploaded this presentation by Charlie Marsh (astral.sh) about how and why he built the popular Python package manager <code>uv</code>. It's great!</p>\n<p><a href=\"https://www.youtube.com/watch?v=gSKTfG1GXYQ\" target=\"_blank\">https://www.youtube.com/watch?v=gSKTfG1GXYQ</a></p>",
      "rawMarkdown": "Jane Street recently uploaded this presentation by Charlie Marsh (astral.sh) about how and why he built the popular Python package manager `uv`. It's great!\n\nhttps://www.youtube.com/watch?v=gSKTfG1GXYQ",
      "votes": null
    },
    {
      "id": "3027001",
      "postDate": "10/24/2024 12:32:44",
      "content": "<blockquote>\n  <p>Uploading the pip wheels as Kaggle Datasets and then using that feels like a very roundabout way of handling dependencies.</p>\n</blockquote>\n<p>Kaggle released something recently to aid in this:</p>\n<p><a href=\"https://www.kaggle.com/discussions/product-feedback/532336\" target=\"_blank\">https://www.kaggle.com/discussions/product-feedback/532336</a></p>",
      "rawMarkdown": "> Uploading the pip wheels as Kaggle Datasets and then using that feels like a very roundabout way of handling dependencies.\n\nKaggle released something recently to aid in this:\n\nhttps://www.kaggle.com/discussions/product-feedback/532336",
      "votes": null
    },
    {
      "id": "3027029",
      "postDate": "10/24/2024 13:05:38",
      "content": "<p>Wow, this is great! Thank you, was not aware of this.</p>",
      "rawMarkdown": "Wow, this is great! Thank you, was not aware of this.",
      "votes": null
    },
    {
      "id": "3027118",
      "postDate": "10/24/2024 14:22:11",
      "content": "<p>I'm looking at their other videos. It's crazy that they're still so devoted to Ocaml.</p>\n<p>I remember working on ocaml for a while many years ago. I loved the language, has anybody tried to run LLM's and ML in ocaml? Is that even possible?</p>",
      "rawMarkdown": "I'm looking at their other videos. It's crazy that they're still so devoted to Ocaml.\n\nI remember working on ocaml for a while many years ago. I loved the language, has anybody tried to run LLM's and ML in ocaml? Is that even possible?",
      "votes": null
    },
    {
      "id": "3027515",
      "postDate": "10/25/2024 00:25:08",
      "content": "<p><a href=\"https://opam.ocaml.org/packages/llama-cpp-ocaml/\" target=\"_blank\">ocaml bindings of llama.cpp</a></p>\n<p>this should be the easiest approach</p>",
      "rawMarkdown": "[ocaml bindings of llama.cpp](https://opam.ocaml.org/packages/llama-cpp-ocaml/)\n\nthis should be the easiest approach",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3027001,
      "author_name": "julianmukaj",
      "author_url": "",
      "post_date": "10/24/2024 12:32:44",
      "content": "<blockquote>\n  <p>Uploading the pip wheels as Kaggle Datasets and then using that feels like a very roundabout way of handling dependencies.</p>\n</blockquote>\n<p>Kaggle released something recently to aid in this:</p>\n<p><a href=\"https://www.kaggle.com/discussions/product-feedback/532336\" target=\"_blank\">https://www.kaggle.com/discussions/product-feedback/532336</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 3027029,
          "author_name": "carlolepelaars",
          "author_url": "",
          "post_date": "10/24/2024 13:05:38",
          "content": "<p>Wow, this is great! Thank you, was not aware of this.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 3027118,
      "author_name": "verracodeguacas",
      "author_url": "",
      "post_date": "10/24/2024 14:22:11",
      "content": "<p>I'm looking at their other videos. It's crazy that they're still so devoted to Ocaml.</p>\n<p>I remember working on ocaml for a while many years ago. I loved the language, has anybody tried to run LLM's and ML in ocaml? Is that even possible?</p>",
      "votes": null,
      "replies": [
        {
          "id": 3027515,
          "author_name": "narainp",
          "author_url": "",
          "post_date": "10/25/2024 00:25:08",
          "content": "<p><a href=\"https://opam.ocaml.org/packages/llama-cpp-ocaml/\" target=\"_blank\">ocaml bindings of llama.cpp</a></p>\n<p>this should be the easiest approach</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3026912": "Jane Street recently uploaded this presentation by Charlie Marsh (astral.sh) about how and why he built the popular Python package manager `uv`. It's great!\n\nhttps://www.youtube.com/watch?v=gSKTfG1GXYQ",
    "3027001": "> Uploading the pip wheels as Kaggle Datasets and then using that feels like a very roundabout way of handling dependencies.\n\nKaggle released something recently to aid in this:\n\nhttps://www.kaggle.com/discussions/product-feedback/532336",
    "3027029": "Wow, this is great! Thank you, was not aware of this.",
    "3027118": "I'm looking at their other videos. It's crazy that they're still so devoted to Ocaml.\n\nI remember working on ocaml for a while many years ago. I loved the language, has anybody tried to run LLM's and ML in ocaml? Is that even possible?",
    "3027515": "[ocaml bindings of llama.cpp](https://opam.ocaml.org/packages/llama-cpp-ocaml/)\n\nthis should be the easiest approach"
  },
  "source": "meta"
}