Llamaindex launches a cloud service to build unstructured data agents

Agents The next big thing in AI. Some define these “agents” differently from others, but the general idea is that they are tools powered by AI that can perform autonomous tasks.

The Hype agent reached a fever, but a startup was relatively early in the game: LlamaindexS Based Former Uber Research scientists, Jerry Liu and Simon Suo, in 2023, Llamaindex allows developers to build personalized agents over unstructured data.

“Llamaindex started as an open toy code project in November 2022,” Liu told TechCrunch. “I became deeply interested in finding out how large language models (LLM) can be used on their own data outside their training set and built an initial set of tools that allow developers to index and include data in their LLM applications.”

Using an open Llamaindex software that has gained millions of GITHUB downloads, developers can create personalized agents that can retrieve information, generate reports and insights and take specific actions. Llamaindex provides connectors and utilities such as Llamaparse, which transform unstructured data into a structured format that can be used for specific AI applications.

Although there are other frames with an open code for building AI agents there, Llamaindex is differentiated by its set of data ingestion, data management and data indexing solutions and retrieval, Liu said. It can link file data such as PDFS and PowerPoint presentations, as well as applications such as a concept and slack, with an agent.

Salesforce, KPMG and Carlyle are among the companies that use Llamaindex today, Liu said.

“All these competitive solutions solve specific problems in different parts of the generative AI stack, but then it is the responsibility of the developer to bring together fragmented solutions to create a working agent,” Liu added. “This is a significant pain that pursues shipping agents to production. Llamaindex has made our mission to provide the most secure, accurate and easy-to-use platform for the construction of end-to-end knowledge agents. “

The next chapter of Llamaindex is a corporate service built on top of the open source company’s offers. Called Llamacloud, it allows customers to create cloud hosted agents that can work and manipulate unstructured data in different formats.

Llamacloud can be implemented by installing software as a service or in a virtual private cloud and comes with features, including role -based access control and single input, Liu said.

In part, to help financing the development of Llamacloud, Llamaindex recently raised $ 19 million in a series of funding, which was led by Norwest Venture Partners, and also saw participation from Greylock. The new money reckoned that the total funding of Llamaindex has been raised to $ 27.5 million, and Liu says it will be used to expand the Llamaindex team of 20 people and the development of products.

“We have enough track to translate us through the initial commercial expansion of our platform,” Liu said. “We are betting on a future in which developers play a big role in providing Genai applications within the enterprise.”