@bogo I did see your talk in the morning and it was brilliant. Are you going to any meetup after #FOSDEM2024 ? I need to buy you a beer to say thank you.
In a surprise twist, our #fosdem2024 BoF session "Special Purpose Operating Systems: The Next Step in OS Evolution or One-Trick Ponies?" was greenlit this morning!
Now again #LLMs: if you do not want to train your own #ai foundation model, you can patch it with so-called #adapters. Benjamin Trim talked about their own open-source adapter micro framework: #Refiners work on top of #PyTorch and use declarative layers to patch models, context API to store state. #fosdem2024
This is one of the most unique and fascinating #Linux talks that I've seen. Happening now at @fosdem in the #Distributions DevRoom, Vojtech Polasek and Lukáš Tyrychtr are presenting the state of #accessibility in the Linux desktop, with an overview of the basics, what the landscape is like, live demos of working accessibility and absence of accessibility, and developing on Linux #a11y tools.
@ffmancera presented in the @fosdem#Distributions DevRoom today about how not to break Linux distros when building a project that several distributions rely on.
Demo time: WiFi is not working, changing to video. Demo gods still in a bad mood at #fosdem2024
There are a lot of good videos on YouTube about doing RAG with Langchain, also with open-soure models. #Langchain#ai#python#fosdem2024
At #fosdem2024 , Jarek Potiuk made¹ an interesting point about why the #AirFlow pipeline manager uses a semi-declarative approach: semantic scalability. When defining complex pipelines, you often end up generating the declaration with a script, which essentially discards the benefit of the approach.
What's fun is that it is the same reasoning I used in my article about #liquidprompt's design², advocating for its procedural theming approach.
Stefano Fancello is talking about #Langchain, an open-source Python-based toolkit for Retrieval Augmentation Generation #RAG. It helps preparing your own data as a context for a question you send to a Large Language Model #LLM. Langchain tools can ingest all kinds of document formats, split documents into Chunks, and create so called #Embeddings and send it to the LLM. #fosdem2024#ai#opensource