Isotopus On Thursday, a healthy million came out of stealth with 20 million badge rounds.
It offers an AI agent to solve a problem in which data analtics products have struggled for decades: People who know how to run big data infrastructure are not those who need to use data.
Through LLMS, business managers can ask questions about their data in natural language. Osotopus agents can provide drafting of Eden, answers and drafts, wherever the finance apps, ERP, CRM, and cloud storage collect data from where the data collects data.
Already there is an offering for business analytics for numerous agents, but there is a unique version of the co -founders of Osotopus. So their product is so sophisticated, Startup has already applied for 10 patents, co -founder CEO Arun Murthy told Tech Crunch.
Just 20 years ago, when the statue was in the mid -20s, he worked in the team in Yahoo, which created an open source project called Hoodop. Hudop promoted the frenzy of the early 2010’s biggest data.
In 2011, Yahoo entered it to a company called Horten Works, which was a co -founder and chief product officer with the statue. Just four years after the launch, Horton Works went publicly. But the rise of New Cloud Storage Tech took advantage of it in the Hudup market, and Horton Works eventually joined with his biggest rival, Clodira. The integrated company was taken private in 2021 after the famous worker investor Carl Icon joined.
The statue went to Claudira for a few years during the uproar, in which around 200 people were arranged. Still, they say, even when he saw the problem of access to ancient figures. He remembers quarter conference calls with Wall Street analysts, who are passionate executives on operating details.
Taxkarnch event
San Francisco
|
October 27-29, 2025
They could not always answer because they did not have data access. “It was shameful,” he admits now. “We were a big data company selling it.”
In 2021, he did not leave the job without realizing what to do next, and a VC introduced it to measure AIA’s Alexander Wang. After some chats and a little consulting with Wang, the statue joined the scale as the Chief Technology Officer.
It was like “to do a PhD on a scale”, he describes. “Understand what these models run and how to improve them.”
But when his old friend from Horton Works called him, Jiachandra called him, both decided to start AI startup himself. He also agreed to join Gopal Vijayraghavan, his third sponsor, from the days of Horton Works, and laid the foundation for Isotops at the end of 2024. Their seeds were led by VAB Goyal in the NTTVC. (Goyal was previously in Norwest Ventures.)
The founders’ background has allowed them to make data from wherever they are stored (whether it be a sales force or a snaplec), but then clean the data. The agent also maintains a lot of context memory to prove useful for complex tasks.
Murthy said, “This is far from a simple chat boot. For example, if Team is asking Eden to prepare a report on repeated revenue trends,” the statistics with Eden are not really present, at least you do not need to chat with you. This is a multi -stop plan, a very complex plan: read, clean, clean, clean and normalize, join data, prevail revenue. “
The agent points to irregularities in his actions, arguments, assumptions and data. Even will make recommendations about ways to move forward. Osotopus has also promised that enterprise users can deploy any of their data to the AI ​​model makers without giving the agent to the agent.
Nevertheless, as it can be as sophisticated as osotopus, startup is in large quantities. The responsible The table of sales force Already offer agents (between the large AI agent of the sales force), and many start -ups with impressive pedigers are also in the market, such as Vesodomai.