News for hackers. Stuff that matters.
Re:LangChain: The Missing Manual by minimaxir on Friday May 19, 2023 @02:18PM (#36005150)
Given the comments in the previous submission about LangChain (https://news.ycombinator.com/item?id=35820931), I am working on a much simpler/faster/cheaper alternative that doesn't require delving deep into arcane documentation and guides as if LangChain's complexity and inflexibility is a good thing for more than quick demos.

Also relatedly, I am also working on a tutorial on how to do vector similarity search without having to pay for a vector store because there is confusingly very few blog articles highlighting the space between "embeddings as text in a CSV" and "full-on vector store management" which is annoying for people who want to do personal projects.

Re:LangChain: The Missing Manual by simonw on Friday May 19, 2023 @02:20PM (#36005167)
Looking forward to seeing that. I haven't yet managed to get my head around LangChain, excited to see what your simpler alternative looks like.
Re:LangChain: The Missing Manual by behnamoh on Friday May 19, 2023 @04:23PM (#36006429)
Always a pleasure to read you two guys' comments on LLM posts!
Re:LangChain: The Missing Manual by minimaxir on Friday May 19, 2023 @04:57PM (#36006768)
it’s not intentional, I promise
Re:LangChain: The Missing Manual by mkw5053 on Friday May 19, 2023 @05:35PM (#36007114)
I used langchain to make a pretty basic LLM augmented with a (free) local vector database: https://github.com/mkwatson/chat_any_site
Re:LangChain: The Missing Manual by danmaw on Friday May 19, 2023 @06:43PM (#36007623)
That's awesome. Looking forward to trying it.
Re:LangChain: The Missing Manual by axpy906 on Friday May 19, 2023 @06:11PM (#36007377)
I would like to see the performance improvement wrt to something like FAISS tuned than using a vector db. Could poke a hole in a budding trend.