scale ai’s former cto launches ai agent that could solve big data’s biggest problem

Isotopes AI Launches with $20 Million in Seed Funding

Isotopes AI officially emerged from stealth mode on Thursday, securing $20 million in a seed funding round.

The company provides an AI-powered agent designed to address a long-standing challenge within data analytics: a disconnect between those who manage data infrastructure and those who require insights from the data itself.

Bridging the Data Gap with AI

Leveraging Large Language Models (LLMs), Isotopes enables business leaders to query their data using natural language.

Their agent, named Aidnn, delivers answers and generates intricate planning documents by accessing data from diverse sources, including financial applications, ERP systems, CRM platforms, and cloud storage solutions.

A Unique Founding Team and Innovative Approach

While numerous agentic business analytics solutions are currently available, Isotopes distinguishes itself through the extensive experience of its co-founders.

The sophistication of their product is underscored by the fact that the startup has already filed applications for 10 patents, as stated by co-founder and CEO Arun Murthy in an interview with TechCrunch.

The Legacy of Hadoop

Over two decades ago, Murthy played a key role at Yahoo, contributing to the development of Hadoop, an open-source project that ignited the Big Data movement of the 2010s.

In 2011, Yahoo established Hortonworks, spinning off Hadoop as a separate entity, with Murthy serving as co-founder and chief product officer.

Hortonworks successfully went public just four years later.

Navigating Industry Shifts

However, the emergence of new cloud storage technologies impacted Hadoop’s market position, ultimately leading to a merger between Hortonworks and its primary competitor, Cloudera.

The combined company was subsequently taken private in 2021 following the involvement of investor Carl Icahn.

Murthy joined Cloudera for a period, overseeing a team of approximately 200 individuals.

Recognizing a Persistent Problem

Even during his time at Cloudera, Murthy observed the recurring issue of limited data access.

He recalls instances during quarterly earnings calls where Wall Street analysts pressed executives for detailed operational information, which they were often unable to provide.

“It was a source of embarrassment,” Murthy admitted, “especially considering we were a big data company.”

From Cloudera to Scale AI and Beyond

In 2021, Murthy departed from Cloudera and was introduced to Alexandr Wang of Scale AI through a venture capital contact.

Following discussions with Wang and a period of consulting work, Murthy accepted the position of chief technology officer at Scale AI.

He described his time at Scale as an intensive learning experience, deepening his understanding of the factors driving AI models and methods for their improvement.

The Founding of Isotopes

A conversation with his former Hortonworks colleague, Prasanth Jayachandran, sparked the idea for a new AI startup.

They then enlisted Gopal Vijayaraghavan, also from their Hortonworks days, as a co-founder, and in late 2024, Isotopes was officially founded.

The seed funding round was led by Vab Goel at NTTVC, who previously worked at NorWest Ventures.

Capabilities of the Isotopes Agent

The founders’ collective experience has enabled them to develop an agent capable of locating data from various storage locations, such as Salesforce and Snowflake, and subsequently cleaning and preparing that data for analysis.

Furthermore, the agent maintains substantial contextual memory, enhancing its effectiveness for complex tasks.

Beyond a Simple Chatbot

“This solution goes significantly beyond a basic chatbot,” Murthy explained.

For instance, when tasked with creating a report on monthly recurring revenue trends, Aidnn must undertake a multi-stage process involving metadata extraction, data reading, cleaning, normalization, revenue proration, and aggregation, as the necessary data isn’t readily available in a chat-friendly format.

Transparency and Data Security

The agent also provides a transparent view of its process, outlining its reasoning, assumptions, and identifying any data anomalies.

It can even offer recommendations for further action.

Isotopes assures enterprise clients that they can deploy the technology without sharing their data with the AI model providers.

Competitive Landscape

Despite its advanced capabilities, Isotopes faces a competitive market.

Established players like Salesforce’s Tableau are already integrating agents into their offerings, and numerous other startups with strong leadership teams are also vying for market share, including WisdomAI.