Rasgo introduces data analytics platform powered by GPT-4

Rasgo introduces data analytics platform powered by GPT-4

Posted on



Startup Rasgo Intelligence Inc. today debuted Rasgo AI, a software platform that uses OpenAI LP’s GPT-4 model to help companies extract insights from their data.

New York-based Rasgo is backed by $25 million in funding. It raised most of that capital through a $20 million Series A round announced in June 2021. At the time, Rasgo’s flagship product was a feature store, a type of tool used to turn business data into a form that can be easily ingested by artificial intelligence models.

The startup’s newly debuted Rasgo AI platform has a different focus. It’s designed to help business users more easily analyze and visualize their companies’ internal data. According to the company, the platform can speed up some analytics tasks by as much as 80%.

Many enterprises store their information in cloud data platforms such as Snowflake. Interacting with such platforms often requires users to write SQL queries, which can be a complicated and time-consuming task. The result is that some advanced analytics projects can require days or weeks of work.

According to Rasgo, its new Rasgo AI platform enables workers to interact with cloud data platforms using natural language prompts instead of SQL. Workers type their prompts into a search bar-like section of the platform’s interface. From there, Rasgo AI uses OpenAI’s GPT-4 model to turn user instructions into SQL code that can run on cloud data platforms such as Snowflake.

The platform can generate data visualizations to explain the results of an analysis. If a user asks Rasgo AI to break down a store’s annual revenues by product, the platform could draw a pie chart that visualizes which product accounted for what percentage of sales. It also generates a natural language explanation of the results.

Organizations may customize Rasgo AI for their requirements using a tool called AI Manager. According to the startup, customizations can be made by giving the platform natural language pointers on how to carry analyses.

A company could instruct Rasgo AI to fetch specific data in response to certain queries. A retailer, for example, might ask the platform to fetch revenue data from both its physical stores and e-commerce website whenever workers enter the query “summarize last week’s sales.” The retailer’s accounting team, meanwhile, could use the AI Manager customization tool to teach the platform how to calculate annualized recurring revenue.

“The largest impediment to self-serve analytics is that existing tools are incapable of providing knowledge workers with data insights without intervention from the data team,” said co-founder and Chief Technology Officer Patrick Dougherty. “To address this, we’ve employed GPT-4 to perform complex reasoning tasks with dynamic objectives.”

Rasgo says its platform doesn’t require companies to make their business records available to third-party AI models such as GPT-4. Instead, the platform carries out analyses by ingesting metadata. That’s contextual information used to describe details such as when a certain business record was created and by what department. 

Rasgo AI supports multiple data platforms on launch including Snowflake and Google LLC’s BigQuery. Down the road, the startup plans to roll out integrations for additional platforms including Amazon Redshift and Databricks Inc.’s Delta Lake.

Image: Unsplash

Your vote of support is important to us and it helps us keep the content FREE.

One-click below supports our mission to provide free, deep and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *