Hacker News new | past | comments | ask | show | jobs | submit login
[flagged] Show HN: AI SQL Copilot LogicLoop – AI to Generate, Optimize and Debug SQL (logicloop.com)
71 points by akisej on May 12, 2023 | hide | past | favorite | 45 comments
Hey folks!

I’m the founder of LogicLoop AI SQL Copilot. If you’re familiar with querying data, you’ve probably spent quite some time manually writing and debugging SQL queries. If you’re a non-technical business user, you will often need to wait and ask engineers to help you write the SQL to pull the data you need. If you’re an engineer, you might be overwhelmed by all these data pull requests from business users.

With LogicLoop's AI SQL Helper Suite, you can ask your data questions using natural language. Ask AI to discover patterns, suggest, write, fix and optimize SQL queries directly on your custom data schema. You can get results on your own data instantly. Once you have your results, you can visualize them on a dashboard or set up recurring alerts and automations. AI makes data more accessible for business users, and faster to work with for engineers/analysts.

Some ways LogicLoop's AI SQL Helper Suite has helped early users: - Business operations teams can find top customers to email and automate outreach - Risk analysts can discover gaps in their fraud monitoring rules to flag more bad actors - Data engineers can fix and optimize long queries to reduce costs

We don’t think this is a panacea that can replace data analysts, but we think this will make data analysis faster and more accessible to more people. Would love for you to give it a try and share any feedback. Thank you.




It's against HN's rules to have promotional upvotes and/or comments in the threads. Especially not promotional comments. I suppose your friends were just trying to help you out, but HN users are acutely sensitive to this and consider it spamming. They're not entirely wrong, either, because the upvotes and comments we're talking about were not organic. We want genuine, curious conversation here, and this is not that.

This in both the site guidelines (https://news.ycombinator.com/newsguidelines.html) and FAQ (https://news.ycombinator.com/newsfaq.html).


Thanks for sharing how you keep the discussion quality high here. We responded to some thought-provoking questions on this thread about differentiation vs ChatGPT, other SQL Copilots, edge cases like more complicated queries etc. that we believe other HN users and makers in this space will benefit from. As I responded in @chatmasta's vouch (thank you!), we're data and software engineering experts, and happy to provide our original perspective on any other questions you have for us. Thanks for your consideration!


The problem is that most of the upvotes and many of the comments on this post were completely inorganic. They were done by people trying to help promote your thread.

I'm not talking about the replies you posted; I'm talking about top-level comments that appeared to be organic and actually weren't.

If you didn't ask friends/fans/users to upvote or comment, or pass a link around (which is usually interpreted as such a request), then someone else did. No doubt they were trying to "help" you. Whoever it was, they need to understand that on HN, it doesn't help, it hurts. It's against HN's rules and we ban accounts and sites that do it. It's also something that the community here is extremely vigilant about and considers spamming.

If you don't want to get banned, and/or flagged and flamed by HN users, you should make sure this doesn't happen again.


This seems like spam. The blog is mostly ChatGPT generated content by authors who have no bio's or profile pictures. The pricing is absolutely out of this world. Particularly if you consider that most of the legwork is done by the OpenAI API anyway.


The submitter is active in the comments and has a long history of submissions related to similar ideas. That seems to indicate to me that this is a real project and an iteration on a premise they've been thinking about for a long time. I do notice even their comments follow a suspiciously GPT-like template with three bullet points. But even if they're using GPT to generate the comments and blog posts, is that really so wrong? Maybe English is their second language and GPT helps them to more effectively communicate their ideas.

I haven't investigated the blog or pricing, but I'm inclined to give them the benefit of the doubt. I don't think this post should be flagged, so I've vouched for it (although I think my vouching powers were nerfed a long time ago, so hopefully someone else will vouch for it too).

EDIT: I actually don't even see the vouch button. Paging @dang


Appreciate the vouch, @chatmasta! I actually did write those comments myself, and think that the 3-point explanation makes for a clear explanation. My cofounder and I both started our careers as software engineers, and I've been in the data space for over a decade.

A quick Google search for LogicLoop will also show you that we're both featured on Forbes 30 Under 30, and are funded by Tier 1 Silicon Valley VCs. https://www.forbes.com/profile/logicloop/?sh=67ca2b047a89


[flagged]


> Are you in on this spam campaign also?

I've had my account since 2012 so feel free to investigate and decide for yourself. I don't know the submitter and have zero affiliation with the product (although my startup does happen to be in a similar space). I'm just sympathetic to someone posting a Show HN and getting flagged based on pure speculation.

Regarding their comment activity: they have comments from earlier this year discussing the same topic (GPT for SQL), and they've previously submitted links to articles and products related to data management.


Stop writing nonsense please. First he is a reputable submitter with a lot of submissions and now all of a sudden it is about you?

Also FYI - there is no vouch button on HN, which hat did you pull that out of?

Oh, and how convenient that your profile is all about SQL also. Just a happy little coincidence isn't it?


You broke the site guidelines extremely badly in this thread. You can't attack other users like this on HN. We ban accounts that do.

If you'd please review https://news.ycombinator.com/newsguidelines.html and stick to the rules when posting here, we'd appreciate it.

p.s. I do appreciate your concern for the quality of HN. but this is not at all the way to advocate for it.


Relax.

> now all of a sudden it is about you?

Am I missing something? You accused me of being affiliated with the person you accused of being a spammer. You made it about me; I'm simply defending myself.

> there is no vouch button on HN, which hat did you pull that out of?

There is a vouch button for flagged comments, and also for some submissions. I can literally see "vouch" links next to [dead] submissions in /new. So I "pulled it out of" my extensive history interacting with the site.

> how convenient that your profile is all about SQL

That shouldn't be surprising to you based on the fact I acknowledged it in my earlier comment. The fact I have a SQL-related startup is the reason I clicked on the Show HN in the first place.


[flagged]


> Who asked you to defend yourself?

You did, when you accused me of being a shill. (Perhaps the resulting toxicity is precisely the reason why there is a rule against accusing someone of being a shill.)


[flagged]


Dude, go for a walk. You sound unhinged. Why are you quoting the definition of gaslighting to me? You're describing your own comments (you accused me of being "part of this spam campaign," so I defended myself by assuring you I am unaffiliated with the submitter, and then you implied this didn't happen by asking why I was defending myself - in other words, you "gaslighted" me. If you question this reality, I suggest scrolling up and reading your comments, all of which I quoted inline in my replies to them).

btw, here's a link [0] to their documentation page which includes a video of the "co-founder of logicloop," featuring her "name and face" (the absence of which you found so suspicious about the blog posts).

It took me two minutes of clicking to find this page and make the determination the site is probably not "spam." If anything, it's a vaporware landing page intended to validate a product that doesn't fully exist yet. And perhaps it shouldn't be a Show HN since you can't immediately try the product. But I think it's definitely unfair to classify the submission as "spam" based on some indicators that don't even apply to it (you say there is no name and face, but I just linked you to a video with the name and face of the cofounder).

[0] https://docs.logicloop.com/


Please don't respond to a bad comment by breaking the site guidelines yourself. That only makes things worse.

https://news.ycombinator.com/newsguidelines.html


[flagged]


> go check out this website

That's what I just did. After clicking for two minutes, I found a video with the author's name and face, which directly contravenes the primary evidence you cited (lack of name and face on blog posts) for the site being spam. The documentation also includes screenshots of the product.

Like I said, I'd be willing to concede it might be a vaporware landing page that is collecting demo leads, and therefore ineligible for a Show HN. But as far as I can tell, the author is a real person and there is at least some work behind the product, as demonstrated by the existence of a website, documentation and explainer videos.


Seems useful, but I'm guessing this will be (or already is) a super competitive space. How is this different than any other AI SQL generator?


Right, here's some ways we're different -

* On your data schema: A lot of the ones we've seen recently don't actually generate these queries on your schema, so you still have to do a lot of manipulation or calculation to get to SQL that just produces results.

* A better experience in productionizing SQL: Like code, writing production SQL is rarely one-shot. LogicLoop comes with a suite of AI SQL Copilot tools, like a fixer, optimizer, editor etc. that allow you to continue to iterate on your SQL logic in an integrated environment.

* Go beyond SQL: Generating query results might often be the first step of a business process. With LogicLoop, you can actually visualize results, send Slack alerts, create tickets, run automations etc. on a schedule.

Hope that helps clarify your question.


The concept looks great. One comment here will be easier flow in the suggestion bit. Instead of letting user input for suggestion on the sql improvement, perhaps doing it with a button click, and with sql automatically populated for each suggestion, without another round trip letting user "ask AI for giving me the queries" in the text input.


That's a great idea for future improvements, thank you for trying it out!


Unfortunately, we had a recent discussion internally amongst the execs and have decided to not allow anyone use tools like this.

There's just no way to know for sure how securely data are stored and what possible attack vectors are.

It's a great idea but imho the only path forward is allowing custom deployments against pretrained models.


According to docs it only takes in your data schema and outputs the SQL query to run, so the AI model doesn't actually get information about your data, just the schema. Otherwise data is handled according to SOC II principles similar to any other BI tools.


Doesn't this penalize the tool though? Knowing the amount of data in a table is still important to determine the type of query, even more to optimize it


Yeah, in general the more data you're able to use (assuming the context window supports it), the better results tend to be. We arrived at the data schema being a good enough compromise at which the benefits outweigh the risks for several use cases. Besides, some data stores that are generated by third-parties actually have common schemas (think Sendgrid / Hubspot activity data), so you're not risking much but potentially gaining a lot of sales ops productivity.


Db schemas sometimes contain important information themselves, as well as proprietary architecture.

I see it same as slack leaking private channel names.


While the marketing video on your website looks like overhyped bullshit, I can see some value in this product. I find that the stupidest bugs in SQL take up the most of my time so if AI can help me fix those faster it will save me time. What’s the accuracy rate on this so far? I imagine it’s not going to be perfect.


Yeah, it's actually pretty good at fixing existing queries if you give it your schema. I'd say on queries from 10-50 lines of SQL it works 9 out of 10 times.


Pretty cool if you have all your business data centralized. But for business users, it could be useful to do something similar directly with APIs of the tools they use (e.g. SFDC, Gainsight, NetSuite, etc.). Is that in the roadmap or are you familiar with another tool that does that?


Yes, we can actually connect to third-party apps via APIs https://docs.logicloop.com/data-sources/supported-data-sourc...


The demo video really just seems like a form input for something I could already just type into ChatGPT myself. Maybe it's because I'm not a LogicLoop user with the integration, but this doesn't seem all that useful nor novel as a standalone product.


Yup, our bet is that people are going to want better integration into data sources and actions resulting from their queries, and that a lot of business value will come from that.


How is this any different from using ChatGPT myself?


Good question. A few different ways:

* Model Selection: We pre-select the best models for SQL generation, so you don't have to A/B test and figure it out. As the number of models available increases, this can take a lot of time if you want to do it yourself.

* Prompt Engineering: Similarly, we've had to fiddle around with a lot of different prompts to deal with common issues like hallucinations (and are constantly updating them, fixing bugs, telling the model what it can't ignore etc.) This also takes a lot of time, patience and skill.

* Directly on your data schema: You'd have to copy-paste your schema every single time you wanted to generate or modify a SQL query, which can be cumbersome. You connect your data sources once with LogicLoop, and it gets auto-included correctly for you.

Hope that helps.


Congrats on the launch! We built something similar a few months ago but couldn't get the quality of the SQL generation to be that good without providing a ton of few-shot examples, especially with databases that contained more bespoke data. If you have enough few-shot examples the quality might be quite good!


Thanks for sharing your experience. By organizing and storing queries, we hope to be able to improve these suggestions as well behind the scenes.


This is great, but how is your performance on multi-table joins? We've been working on a homebrew solution using OpenAI internally conditioned on our schema, and can't bridge the multi-table gap.

Also I'm skeptical if this generalizes, do you have measures in place to prevent query hallucination?


On simpler multi-table joins we've been able to product good results, and we've done a lot of prompt engineering to make sure it takes the schema very seriously so that prevents hallucinations too. We're always finding new edge cases and fixing those as we go.


Is this a custom model, or based on OpenAI's API?


We use a combination of APIs from existing LLM providers, and do some serious prompt engineering to get the best from them. We're starting to train models on more SQL-specific prompts now.


Do I understand right, everything goes to your server and all the processing happens there rather than on-premise? Too bad, that pretty much ensures no serious company will even touch it.


I understand your skepticism, here's how we think about things -

* We only send data schema to underlying models: No data is actually sent there, but yes when you connect your database, we run the generated SQL on your database and store results to show those to you. They get regularly deleted though.

* We take security seriously: We're SOC2 Type 2 compliant and publish our security practices online, have a DPA, etc.

* Companies value convenience: We've often seen that companies _prefer_ cloud deploys so they don't have to DevOps themselves. Can be a matter of personal preference.

Hope that helps clarify.


This is really neat. Do I need to give access to my data or just the schema?


Thanks! We're starting with just the schema, good clarification.


[flagged]


Haha, almost :P Some problems can't be solved. For everything else, there's LogicLoop.


[flagged]


It varies. Theoretically, someone who doesn't know SQL should be able to generate it from this, but you'd still have to get someone to connect your datasource, and have a general sense for what you want to do.

I'd say it's not optimized for the high-level executive or entry-level operations person. It's more geared for someone that is data-aware but wouldn't consider themselves confident in SQL, like a junior data analyst profile.


[flagged]


Thank you, let us know what you think when you try it out.


[flagged]


Agreed, that's always a risk with making data too easy to access, its context doesn't always come with it. While we initially thought it'd be most valuable for new SQL users, a lot of existing SQL experts actually find that it gives them a huge productivity boost.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: