RD-11.15-16.2022-Einblick

BENEDETTO BURATTI: All right. Good afternoon, everybody. My name is Benedetto Buratti, and I'm one of the co-founders, head of data science at Einblick. I'm here in place of our CEO, Emmanuel Zgraggen. He was a postdoc at MIT CSAIL. And actually the whole company is based out of MIT researchers, I will talk about in a second.

So Einblick is a web-based data science platform. So it's designed for Python users, SQL users, and it's really designed to make their work more productive and collaborative. It's based on more than 60 years of research at MIT and Brown University. And the whole team essentially is based out of academia.

So what Einblick is about, so how Einblick started. So the whole idea of Einblick really starts from trying to demystify a little bit the Hollywood-given idea of analyzing data, right? If you think about movies like Minority Report, you have a smart guy that is in front of a big screen trying to move things around and somehow achieve the result really fast, really quick, right? But the folks in the audience that work with data know that this is not actually how it works in practice.

And so if we had to break down in three big buckets what are the three different categories of data analytics platform, data science platform, on one side you have the workflow engines, like Alteryx or NIME, where really our focus is on making more accessible complex workflows, right? Great for make data science more accessible, but a little bit how our time in terms of collaboration.

The second bucket of tools are BI tools, your Power BI, for example. Great tools for dashboarding but really limited in terms of data manipulation, on a general sense. And then you have your traditional Jupyter Notebooks, the last bucket, right? You can think of your data scientist that can sit on a computer, write code, and essentially work with the data directly. A lot of expressiveness there, a little bit limited in terms of collaboration and ease of access.

So when we were designing Einblick, we were really focusing on overcoming those limitations, right? So eliminating repetitive tasks and tedious tasks that are really common among Jupyter users, and then just make data science more collaborative, more easy, more accessible. So if you're familiar with Jupyter Notebooks, if you're a data scientist or an analyst, you might be familiar with Jupyter Notebooks. But also if you're familiar with Miro boards, idea boards, or collaborative tools like Figma, you can think of Einblick as a combination of those two worlds, right?

And so this is how Einblick looks like. It can work in person on those large interactive monitors. And we have a booth here and you can play with this kind of version in real person. And the other version where I really think Einblick excels is actually the remote online collaboration version. So the research project that I was talking about before was really focused on making this large screen more interactive and analyzing data through those screens.

When COVID hit, a lot of people started working remote. So that model was not that interesting anymore. And so we completely switched to offering online offering. So just think about your analyst team, your data science team, joining one Canvas. They can lay out different operators, bring in different data sources. They can connect to simple CSV file or database systems and collaborative work on data problem.

And again, when I say collaborative, I don't mean that you have to collaborate. If you want to play single player mode, you can. But if you like multiplayer mode, you can definitely do that in Einblick. So just to give you an idea of how Einblick has been used, I want to just talk about, real quick, on this use case from a leading German chemical manufacturer.

So the setup there was that this company had a big engineering culture, of course. But it was not a data science culture, of course. So the analytics and data science team was really focused on working, on coding their models, and using R. So their struggle was being more productive, like being able to react faster to the different things that will happen to the plant, and to have visibility with the management, the stakeholders, right? They're building these really sophisticated models, but then people have a hard time understanding what's going on.

And so with Einblick they did a parallel experiment. They used Einblick and R to solve the same use case. They were able to deploy a model in less than a week using Einblick, and the other team essentially took a month to achieve the same result. And at the same time, while using Einblick, they were able to share the results really fast.

So as a last note, I just want to say that Einblick's a virginal purpose platform, so we can serve different verticals. The main ask that it's for this conference and what we are really interested in is to talk with data scientists, analysts, anyone that is working with data. So if you work with data and you are interested about Einblick, you are interested about collaboration, stop by the booth. We are here on the left.

We also prepared a small gift for you, and so that you can start using Einblick and you can experience the whole platform in all its features. So thank you very much for your attention and talk to you later at the booth.