deepset, the mission-critical AI company, is unveiling deepset Studio, a free, user-friendly enabler of custom AI development that abstracts the rampant complexities from AI pipeline development. With a visual, drag-and-drop environment, engineering and data teams can create robust, composable AI systems with ease—including agentic and retrieval augmented generation (RAG) apps—which are immediately deployable within cloud and on-prem environments via deepset Cloud and NVIDIA AI Enterprise software.
AI teams—across industry and enterprise size—are dealing with the fervent push for AI applications that deliver tangible value. To achieve this, teams are stitching together a variety of components, models, and data, creating deeply custom—and simultaneously, complex—AI pipelines that reflect their industry’s unique needs. These compound AI system architectures, though tailored, are an amalgamation of intricacies, creating difficulties in maintenance and management.
deepset seeks to radically redefine how engineers and data teams craft AI pipelines, streamlining the process without sacrificing on customizability, according to the company. deepset Studio serves as a visual programming interface that allows AI teams to “create and debug these compound AI systems while keeping cognitive load at a minimum,” noted Jay Wilder, VP of marketing at deepset.
With deepset Studio’s intuitive interface, enterprises can simplify and accelerate the AI development process while still being able to architect for a variety of industry-dependent needs and large language model (LLM) use cases, from agentic to RAG and natural language processing (NLP).
deepset Studio’s drag-and-drop visual editor automatically validates component relationships and pipeline structure, underpinned by popular open source framework Haystack's extensive library of integrations and components. The solution also offers proven pipeline templates, component configurations, and shareable visual representations so that AI teams can easily jumpstart the development process, according to deepset.
“The idea here is to…[take] the [AI] space—which keeps getting more sophisticated and complex—[and] keep simplifying it and making it more and more powerful for our users to take advantage of all this innovation,” said Wilder. “One of the most important parts is that it is free to use, so there's really no barrier to entry.”
Additionally, “everything that you build in Studio is one-to-one with the actual code in Haystack,” explained Wilder. “As you move components around, if you change them, it’s immediately updating your code…you can take that at the end and run that wherever you like…in the environment of your choice,” including deepset Cloud and NVIDIA AI Enterprise.
deepset Studio is offered as a free standalone tool for users of Haystack as well as integrated with NVIDIA AI Enterprise for cloud or on-premises deployments of production AI. deepset Studio is integrated with NVIDIA NIM microservices and the NVIDIA API catalog, allowing users to configure NIM deployments and LLM inference from directly within Studio.
“Enterprises across industries are seeking ways to effectively integrate AI into their core operations while maintaining security and scalability,” said Anne Hecht, senior director of product marketing for enterprise software at NVIDIA. “The integration of the NVIDIA AI Enterprise software suite with deepset Studio will help simplify and accelerate the deployment of AI applications, supporting both cloud and on-premises environments.”
To learn more about deepset Studio, please visit https://www.deepset.ai/.