DataStax snaps up Langchain-democratiser Langflow
"You have to chain user input, system prompts, and DB data to feed the LLM and then lots of processing to deliver that magic AI agent experience to the user"
DataStax has acquired Langflow, which provides a slick GUI for building retrieval augmented generation (RAG) and other AI applications with LangChain, as it continues to build out its generative AI capabilities.
It said that the startup will “continue to operate as a separate entity, focusing on product innovation and community collaboration.”
Terms of the deal were not disclosed.
Maturing generative AI toolchains
Many enterprises looking to deploy generative AI in production are heavily relying on RAG to ensure that models are informed by business data that may not have been available to them in training – and as a result can furnish more specific responses with much less hallucination.
Even with the popular LangChain framework this process could be a lot easier; the generative AI toolchain is still immature and often unwieldy.
See also: Avoiding “RAG hell”
Langflow, which has over 15,000 stars on GitHub, aims to make it far easier to build generative AI applications using a drag-and-drop visual environment and reusable components for multiple apps or data sources.
Datastax is best known for providing a managed enterprise Apache Cassandra-based DBaaS, Astra DB, as well as a Pulsar-based data streaming service Astra Streaming; it has been working extensively with clients on enterprise AI deployments over the past twelve months.
With the addition of the startup it will be able to provide a “one-stop generative AI application stack, enabling easy development with drag-and-drop tools and leveraging partners like LangChain and LlamaIndex” as well as integration with Astra DB DBaaS “for unified data access and global-scale deployments” the company said today.
Join peers following The Stack on LinkedIn
As DataStax CPO Ed Anuff put it: "We've spent the last year helping hundreds of companies build Gen AI apps using AstraDB and RAG. Today we made a big step forward in making the one-click RAG AI Stack a reality by acquiring Langflow to bring visual editing of AI flows dead simple.
"If you haven't been paying close attention to how AI apps are built, there's a lot more to it than the LLM. You have to chain user input, system prompts, and DB data to feed the LLM and then lots of processing to deliver that magic AI agent experience to the user. Lots of iteration and lots can go wrong. Langflow makes this super easy!"
Chet Kapoor, DataStax CEO, said. “Langflow is focused on democratizing and accelerating generative AI development for any developer or company, and in joining DataStax, we’re working together to enable developers to put their generative AI ideas on a fast path to production.”
"Langflow's visual framework is a revelation” said Jan Schummers, senior software engineer, WinWeb in a canned statement.
“Langflow simplifies wiring up components, and with its prebuilt connections and visual tools, we can experiment and iterate with unprecedented speed. This will transform our RAG application development, letting us focus more on creativity and less on complexity.”
Rodrigo Nader, CEO, Langflow said: “With DataStax, we will be fully focused on the execution of our product vision, roadmap, and community collaboration, and will continue to add to the greatest breadth of integrations across different AI ecosystem projects and products – including more data sources and databases, models, applications and APIs.”
