How to think about LLMs for OSS development
Briefly

The discussion emphasizes the need to think about fundamental principles rather than current trends in rapidly evolving technologies like LLMs. With constant changes in technology, it is important to focus on basics such as algorithms, data structures, and general decision-making frameworks rather than specific solutions or current 'how to' guides. The aim is to encourage questioning and facilitate meaningful conversations while recognizing that solutions already exist even if they are not widely known.
This is a 'how to think about' talk, not a 'how to'. There are plenty of 'how to's' out there—I want to seed some good hallway conversations and offer a call to action.
The space is changing too rapidly; if I say 'we need a solution to X', chances are someone is already working on that, unknown to me.
In the 90's, the big battlefield was programming languages. Rather than focus on fads, we focused on fundamentals like algorithms and data structures.
By treating LLMs as an unknown value, we can discuss principles that will likely remain relevant and help you make informed decisions about their application.
Read at Boom! Michael Droettboom's blog
[
|
]