Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Dataherald AI – Natural Language to SQL Engine

github.com

Post reply on HN
Print view View original

91–100 of 108 posts

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#91

Post by saigal »

canvascritic wrote: post #39
aazo11 wrote: post #37

Earlier quoted context omitted.

SQL is definitely an old technology. However it is still the main language for interfacing with structured data, even for newer tools like Clickhouse. Down the road it is conceivable that it will be replaced with something else, but the cost of migrating existing data to those will be so high I imagine no-one will take the step to proactively move.

SQL makes the most sense if humans are going to be programming deterministically, but if we're talking about natural language input I can't help but think that compiling natural language to SQL feels a little shortsighted as a long term strategy. Surely it makes sense for a MVP, but I'm curious about how much we're potentially forfeiting in terms of innovation by anchoring so deeply to it with all the possibilities a…

Do you mean that SQL will significantly fade away?

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#92

Post by itsoktocry »

neural_thing wrote: post #34

When I look at all the LLM SQL tools, I think: what a cheap and accessible way to get the wrong answers. SQL is easy. Knowledge management is hard. Does the LLM know that there was a bug in June that changed the data? Does it know that this one column is improperly named and confusing? Does it know that you recently released a mobile app the data from which is in a different table? No, of course not, those things are…

>Does the LLM know that there was a bug in June that changed the data?

No. Does "the SQL"? Also no.

That requirement falls on the person using the tool, doesn't it?

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#93

Post by saigal »

vekker wrote: post #73

This is interesting. I have an enterprise customer who's looking for exactly this. Any plans to support DB2?

happy to connect and describe more to see whether we could collaborate. please feel free to email be at anuj at dataherald dot com or sign up here and we will get in touch https://www.dataherald.com/contact

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#94

Post by Nick_Molter »

neural_thing wrote: post #34

When I look at all the LLM SQL tools, I think: what a cheap and accessible way to get the wrong answers. SQL is easy. Knowledge management is hard. Does the LLM know that there was a bug in June that changed the data? Does it know that this one column is improperly named and confusing? Does it know that you recently released a mobile app the data from which is in a different table? No, of course not, those things are…

Love this! I don't think you could be more right about the practical challenges of implementing something like this. In my experience, this same problem is what makes it so challenging to onboard new data scientists/analysts.

It takes a lot of training to get a team member up to speed - with the same amount of training, do you think an LLM can compete?

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#95

Post by Nick_Molter »

itsoktocry wrote: post #92
neural_thing wrote: post #34

When I look at all the LLM SQL tools, I think: what a cheap and accessible way to get the wrong answers. SQL is easy. Knowledge management is hard. Does the LLM know that there was a bug in June that changed the data? Does it know that this one column is improperly named and confusing? Does it know that you recently released a mobile app the data from which is in a different table? No, of course not, those things are…

> Does the LLM know that there was a bug in June that changed the data? No. Does "the SQL"? Also no. That requirement falls on the person using the tool, doesn't it?

Is there a reason why an LLM can't learn these nuances the same as a human?

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#97

Post by BenderV »

Congrats for your launch !

I also built NL2SQL solution (relying on OpenAI).

My first version was a direct NL2SQL version, named Olympe - https://github.com/BenderV/olympe I used it quite a while but trying to plug it to real database (100+ tables, unprepared) was unsuccesful.

I switched to chat version, named Ada - https://github.com/BenderV/ada

IMHO, it's the way to go. The AI explore the database, it's connection, the data format & co. Plus, it help with ambiguity and feels more "natural".

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#98

Post by Nick_Molter »

BenderV wrote: post #97

Congrats for your launch ! I also built NL2SQL solution (relying on OpenAI). My first version was a direct NL2SQL version, named Olympe - https://github.com/BenderV/olympe I used it quite a while but trying to plug it to real database (100+ tables, unprepared) was unsuccesful. I switched to chat version, named Ada - https://github.com/BenderV/ada IMHO, it's the way to go. The AI explore the database, it's connection,…

Nice! how's the switch to Ada been? Any unexpected hurdles?

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#99

Post by MrezaPourreza »

sandGorgon wrote: post #64

this is very interesting. just curious - is the context store specific to the table structure of the target database ? or is it a very generic one. im asking this cos the DIN-SQL algorithm seems to be a chain-of-thought kind of algorithm (with query decomposition). I'm wondering where do u plug-in the context store ? in DIN-SQL the schema linking module is random samples from Spider. Is that whats in the context stor…

Thank you for your interest in our work. The schema linking approach employed in our agent significantly differs from the one described in my paper. In the paper, we utilized a method that involved breaking down questions and matching entities with the table schema. However, when dealing with large datasets, the approach outlined in the DIN-SQL paper is no longer feasible, as we cannot feasibly provide all table names and columns within the prompt. As a result, we opted for an alternative method, leveraging table embeddings along with the NL question embedding to facilitate the schema linking process.

Regarding the context store you mentioned, it plays a vital role in our system. We use the context store to retrieve previously verified questions from earlier interactions. This functionality assists the agent in locating the correct tables and columns based on the knowledge gained from past interactions with our system.

Re: Show HN: Dataherald AI – Natural Language to SQL Engine

#100

Post by saigal »

cloudking wrote: post #51
aazo11 wrote: post #42

Earlier quoted context omitted.

Depends what you mean by 'work.' You can definitely use them, but GPT-4 is the only model that generates good SQL without lots of training + fine-tuning (and it is very difficult/impossible to come up with Nl SQL training sets). So GPT-4 is the only model that works well in terms of accuracy for now. That being said, I am pretty sure the self hosted models will get there very soon. Are you looking to deploy a solutio…

Yes, we have use cases that can't leverage GPT-4 for data privacy reasons. I was curious if any self hosted models would suffice and could be integrated.

GPT4 is best right now for all the reasons you likely already know. We are working on self-hosted but for right now it is GPT4. We can let you know when we work on self-hosted, which will no doubt come in the future. feel free to email me at anuj at dataherald dot com or sign up on our site (https://www.dataherald.com/contact) and i will keep you posted.

i should note that our (dataherald's) product only sends schema to the LLM, which is important to note.

Post reply on HN