Meetups/Infra/2025-01-27

From Noisebridge
Jump to navigation Jump to search

https://www.noisebridge.net/wiki/Meetups/Infra/2025-01-13#Takeway_requests https://meet.jit.si/nb-meetup-infra


(Preamble:

   = Meetup - Infra =
   https://www.noisebridge.net/wiki/Meetups/Infra
   https://www.noisebridge.net/wiki/Meetups/Infra/2024-..-.. )

2025-01-27m Meetups/Infra

Noisebridge | About | Visit | 272 | Manual | Contact | Guilds | Resources | Events | Projects | WGs | 5MoF | Meetings | Donate V · T · E
Events | Hosting | Streaming | Meetup | Upcoming Events | Anniversaries | Hackathons | External Events V · T · E
Meetups / Infra: 2025 | Template | Pad | Jitsi V · T · E

(TODO summary)

Introductions[edit | edit source]

  • [name] - [background]. [goals for meetup, or interests to explore]
  • Loren --

welcome to ML/LLM themed night of the infra meetup. Tonight, I want us building ourselves workflows with llm tools. A particular workflow -- something for hosting meetings in person. Sharing links, chat, record/ transcript locally.

  • Doug -- today, actively building out workflows for computer vision
  • Wolf -- using an LLM to rewrite my resume this week. (curious about motiviation & success
  • Greg -- nix. interested in running generative AIs locally
  • Kevin -- on his way to nb :) :) -- experimenting with fine-tuning & rag. Have a vector store, spidering. Using clickhouse, sqlite query interface -- lots of magic baked in. Common crawl
  • Sameer -- currently ambivalent about software.
  • Jams -- downloading ollama now. Looking forward to asking questions I wouldn't ask only,
  • Mike -- Software engineer. I like to watch.


Lesson or Demo[edit | edit source]

  • Read aloud: clarify for meetup. We are taking notes in a riseup pad (or I am--help appreciated, and links). We have meeting notes posted to the wiki. noisebridge.net, search Infra, or Meetups/Infra. (the Infrastructure page has a disambiguation link.)
  • Shell, web services, self-hosting, networking!


= Doug -- image recognition[edit | edit source]

difficult task. Identify text from ads & flyers, alongside images. https://github.com/VikParuchuri/surya

tesseract -- likes a column, line, word, whole page -- but

structured outputs with multimodel inputs, it's amazing.

DeepSeek papers[edit | edit source]



Takeaways[edit | edit source]

  • Doug -- downloading surya, thanks Sameer!
    • Kevin & Jams -- vision & text. llama vision.
  • Greg -- has ollama and llm installed now, pulling, playing

https://simonwillison.net/2025/Jan/20/deepseek-r1/

  • Kevin -- more complete version of common crawl vector store for RAG-backed app



Questions, Discussion, or Coworking[edit | edit source]

  • [Issue]

For next time[edit | edit source]

Questions[edit | edit source]

Readings & Exercises[edit | edit source]

  • Readings
  • Exercises


    1. Installing llm using the Python uv tool

After having installed the `uv` tool run `uv tool install llm`

Then you can run it with `uv tool run llm *args`

Join online[edit | edit source]

  • Try it yourself!
    • Join libera.chat #nb-meetup-infra

https://www.noisebridge.net/wiki/Meetups/Infra