Hacker News Reader

lolpandaon OpenAI compatibility

The compatibility layer can be also built in libraries. For example, Langchain has llm() which can work with multiple LLM backend. Which do you prefer?

lolpandaon Show HN: Natural-SQL-7B, a strong text-to-SQL model

I don't think any of those text-to-sql models are solving the right problems. The hard part is not syntax or I don't know how to write a group by query. Most data scientists and engineers spend more time on understanding the meaning of the data. One cannot simply look at a 50 columns table in Snowflake and guess what columns are by their names. For example, we have 10 columns in one tables, all named ...price. We have to go to wiki to find the actual meaning or read the DBT definitions. I cannot trust any queries that models produce because they don't understand the data; they only understand the query syntax.

lolpandaon SeaweedFS fast distributed storage system for blobs, objects, files and datalake

For companies hosting their entire infra on AWS, what's the advantage of SeaweedFS running on a fleet of EC2 machines over storing on S3?

lolpandaon Unattractive funds managers outperform funds with attractive managers by over 2%

link to non paywalled version of this https://d1e00ek4ebabms.cloudfront.net/production/uploaded-fi...

this paper also defines good looking and how to measure it with machine learning algorithms. given that look is highly subjective, any findings based on that is not very useful

lolpandaon Ask HN: Is anyone else bearish on OpenAI?

does llm really have a moat? you may compare google and bing. they provide essentially the same set of functionalities and nearly identical user interface. the users should be able to move from one to the other easily. but google has better search results, at least in users perception. most users don't even get to try the competitors products. openai is in google's position for now. i tried claude a few times. it seems more capable than gpt3.5 but i found myself always come back to chatgpt

lolpandaon Introducing Adept Experiments – use AI workflows to delegate repetitive tasks

this seems useful based on the fact that software pieces do not work with each other. so the human has to manually move data from one to the other. in most of the cases if users always have to do A->B, does it make more sense to build automation in code instead of using ai? the automation can be built by engineers who are also assisted by ai.

lolpandaon Is anyone using Walrus operator? Match statement?

If you're a Python developer, do you use the new language features? If not, why do you not use them?