5.5.22-Efficient-AI-Prescient

ANDY WANG: Hi. My name is Andy Wang. I'm the founder and CEO of Prescient. So most of the startups here today are in the space of data science, which is about building the optimal AI models. We are in the space of data engineering, which is about data preparation, data transformation, feature extraction, and everything that you have to do to prepare your data for AI and analytics.

And as it turns out, 80% of your data work is actually all data engineering. And because data engineering is so complex and so time-consuming, less than 5% of [INAUDIBLE] data is actually used for analytics today. So producing high-quality, consistent data is really the single most effective step for you to create value out of data. And that's what we do.

So we have a distributed data engineering software. In the cloud, you can create this functional block diagram type of data flows that allows you to acquire data, clean data, transform data, extract features, store features, store data, whatever you want it to do. And you can execute your data flows in the cloud or at the edge, at the edge meaning in your plants, out in the field, in thousands of places where you wanted to process your data.

So having that software in the cloud allows you to have a single pane of glass to manage all of your data operations across your organization. But having the ability to execute your data flows everywhere down at the edge enables you to scale to a very large number of data sources across your organization. So that's a very powerful framework for working with data.

So we work with some of the most complex data problems out there. This one here, we work with a tier one automotive equipment provider, as well as their customers, to manage the large amount of data and data sources in automotive manufacturing plants. In manufacturing plants, today you're going to have thousands of data sources, thousands of different models.

All of those are different, and they're actually changing all the time. And there's problems in those data as they come up. So we automatically transform, detect errors, fix things, validate models, so make that very easy for manufacturing companies to homogenize the high-quality, consistent data that they need to perform the next step of data analytics.

Another area where we're getting a lot of traction is working with companies that are deploying AI, in particular deploying AI to the edge. And as I mentioned, industrial data has many different data sources, many different data formats. So if you're AI company, you probably know that just building one single AI solution cannot serve all of your customers or all of your use scenarios.

So what we do is we can help with acquiring that data from any data source, transforming the data from any data format, and feed it into your AI engine. We can also take the output of the AI and do further post-processing, data integration, adding other functionalities to it. AI is never the end result, right? The AI is a big part of your solution, but we kind of help you to customize and extend your AI solutions. We find traction here with Fortune 500 companies, as well as AI startups.

Our product is in production. So we are looking for customers and partners. If you have large, complex edge data problems, many data sources, unreliable data sources, or if you are trying to build AI, but you find that there is just many different scenarios you have to deal with and you don't want to code, produce a dedicated solution for every single case, that's the kind of problems we solve. So if you have those type of challenges, we're happy to talk to you. Thank you.

[APPLAUSE]