This is, according to the website, "a reference architecture for the LLM app stack. It shows the most common systems, tools, and design patterns used by AI startups and tech companies." Basically it breaks down into a four-part structure: data, prompt/examples, query, and output. I think this is just an early stack, and I expect a lot more to be added over time at the top and the bottom. A note about process: this item came into my inbox as part of a larger item from Andreessen Horowitz (a16z). I hesitated to link to it, because of comments I've read critical of the venture capitalist company. But then I saw an Alan Levine post linking to the item, and I changed my mind. It's a lesson to me about the need to be aware of the different factors swaying whether or not I include an item in this newsletter. My initial instinct, I think, was wrong, and I need to be careful about who influences me and why.
Today: 2 Total: 23 [Direct link] [Share]
