deepset has launched deepset Studio, a visual programming interface for its Haystack open-source framework. Studio allows teams to visually architect custom AI pipelines for use in mission-critical business applications. Once built, applications can be deployed using deepset Cloud or on-premise, giving customers a choice of where they want to run them.
Milos Rusic, CEO and co-founder of deepset said, “deepset is a leader in enabling custom AI development – powering many of the world’s most trusted, high-value use cases.
“The addition of deepset Studio now enables developers at thousands of companies worldwide to architect the next generation of custom LLM applications. This new tool, combined with native integrations to NVIDIA, provides the most robust platform for enterprise developers to safely and reliably develop mission-critical Generative AI products and features.”
What is deepset Studio?
deepset Studio is a free visual pipeline design tool aimed at anyone who wants to design AI models. The company has removed users’ need for deep technical and coding knowledge by making it plug-and-play. All they require is an AI use case that they want to implement using an LLM to which they have access.
Studio sits on top of the open-source AI framework Haystack. Users get access to an extensive range of components to kick-start their work. An active community also provides a wide range of free components. Users with knowledge of Python can also build and add their components.
Components can also be used to build pipelines. This is where deepset Studio is adding functionality. It allows users to drag and drop components to create pipelines to speed up AI development.
According to deepset, Studio adds the ability to:
- Design AI pipelines with a drag-and-drop visual editor that automatically validates component relationships and pipeline structure.
- Leverage Haystack’s comprehensive library of integrations and components to create flexible and composable application architectures like RAG and Agents.
- Jumpstart the development process with proven pipeline templates, component configurations, and shareable visual representations of simple to complex AI systems.
- Go to production faster with native cloud and on-premise deployment options for deepset Cloud, NVIDIA Enterprise AI, and the AWS Ecosystem.
In addition to this, the company has integrated Studio with NVIDIA NIM and API catalogue. That means that NVIDIA Enterprise AI users can take their applications and optimise them through deepset Studio.
Enterprise Times: What does this mean?
Building AI-enabled applications is a key pressure point for many development teams. However, most tools come with a steep learning curve, which developers need more time to get to grips with. deepset has set out to reduce that learning curve by providing, through open-source tools, a simpler way for organisations to add AI to their business.
While it doesn’t use the phrase, this is effectively a no-code/low-code approach to AI. You can drag and drop in a no-code approach or script additional components as in a low-code approach. More importantly, it frees up IT development teams and data scientists to build and optimise LLMs while the business builds the apps it needs on top of those.
Haystack has been growing quickly over the last year. With deepset now delivering its Cloud and Studio products on top of that, it should grow even quicker. It will be interesting to see where this goes over the next few months.
