Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Show HN: ChatGPT for your Database. a new way to interact with structured data (chaturdata.com)
14 points by aiassist on Feb 12, 2024 | hide | past | favorite | 7 comments
what it can do for you lately:

let's say you work at uber. uber probably has a database with a bunch of rides data in it. it probably has a rides_fulfilled table, a users table, a drivers table, and a whole lot more tables but just imagine it was that simple for now.

let's say you are in charge of the Atlanta geo and you want to summarize the drivers / riders / financial performance of your city on Superbowl Sunday.

before chat ur data, you'd probably login to the corporate bi tool, change some filters on a dashboard, take some screenshots, go to the next dashboard, take some more screenshots, write an email talking about how much your team crushed it this year and how the game going to overtime helped spread out the demand for pickups over a wider time span (which gave your drivers more time to get on the road), attach the screenshots even though they might not be able to totally reinforce your storyline at all, and send it to your boss who's probably the GM of the Southeast region.

after chat ur data, you send a prompt to your AI Assistant to analyze the data for you compared to the last 3 Superbowl days and summarize it into highlights and lowlights, plus action items you want to implement for the next big event day.

the AI will run some queries on the rides_fulfilled table filtered to your geo, join it to the users and drivers tables to find interesting trends by those dimensions, identify the positive and negative trends, and make some charts that help to visualize those trends + summarize the storyline. while you watch the halftime show a few more times because Usher is your hometown favorite and you're nostalgic for the early 2000s.

you review the storyline + charts, maybe tweak it a little bit, and then send it



>let's say you work at uber. uber probably has a database with a bunch of rides data in it. it probably has a rides_fulfilled table, a users table, a drivers table, and a whole lot more tables but just imagine it was that simple for now.

What I don't get from the landing page clearly is what data will be sent to your app. Will users give you db credentials so that you have full access to schema and underlying data? And later these data will be sent to openai?


sorry about that!

yes, users have to provide db credentials with read-only access to the data.

no, db credentials are never shared with OpenAI. our API handles user authentication and credential storage and calls out to OpenAI for the AI function calls


well, I'm not sure how often people are comfortable sharing db credentials.

Will db data (not credentials) be sent to open ai?


Yes, query results are shared with OpenAI.

We work with lots of companies that are comfortable with it but have met others that say "No Way in Hell". For some reason, they tend to be more comfortable using OpenAI APIs on Azure than OpenAI directly though.


.. and in the process make a few mistakes, make up a number of ten. So later on, when this is discovered, you will be pretty embarrassed.

Using AI to write queries/configs that you later manually review? Good idea. Using AI to query the data and trust its results unconditionally? Terrible idea (unless no one cares about your result)


100% agree. we always show you the SQL that the AI generates so you don't have to trust it blindly! the sad part is when you do manually review it and still miss the issues... i've done that a few times


Reviewing vs creating are very different mindsets. There’s a red team, blue team flavor to it.

AI levels up the less skilled, but makes even the more skilled miss embarrassing mistakes.




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

Search: