Rahul Sonwalkar: recommendation

Transcript context

…here are data scientists? think everyone here needs a data scientist. But there just aren't enough. And that's what we're building. Julius is an AI data scientist that helps you analyze datasets, make visualizations, Get insights from the data, and really dive deep into all sorts of data that we have in real life. So, we launched about six months ago, and since then have grown to 300, 000 users several thousand users using us daily to analyze datasets, create visualizations and get insights. So what I'll do now is give you guys a quick live demo of how it actually works in IA. I actually hope it works because we just posted code changes. But here I have a dataset of 20, 000 rows of data over time for the last 100 years of human height for different countries. So I'm going to take this dataset, dump it in Joly's and say, load this for me. And while it's doing that, I want to explain what's happening under the hood. So basically, for each user, Think about how a human data scientist would analyze a data set that you give it. It would take its computer write code, run that code, maybe in a Jupyter notebook, look at the output, and then decide if that answers your question, or if you need to write more code. Julia works similarly. So that's you, that's the AI, and then for each user, you get a virtual machine in the cloud, and Where the AI is filling up the Jupyter Notebook, writing the code to get the analysis that you want, and then serving that back to you. Many times, that code is not correct the first time. But Julia is able to recover from those errors and actually get you the answer that you want. So let's look at our chat. We said, load this file for me, and the AI basically went, spun up a Jupyter notebook, loaded pandas, looked at the file, and gave us a few rows. I'm going to ask plot the Mail, pipe, overtime, in France.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.