Israeli artificial intelligence dataset startup Dataloop Ltd. today announced a new collaboration with Nvidia Corp. to integrate the Nvidia AI Enterprise software platform into its own platform to advance the building and refinement of AI applications for businesses.

Dataloop argues that though generative AI could automate up to 70% of all business activities, the road to AI adoption can be difficult, with many companies not making it out of the proof-of-concept or pilot stages. Disorganized and disjointed data is a significant inhibitor to progress, whereas facilitating efficient AI development backed by well-managed, optimized data is the key to unlocking AI’s value. This is where the new collaboration steps in.

The provision of Nvidia technology of the Dataloop platform is said to deliver customers leading AI tools directly on a unified, orchestrated platform. The integration of numerous tools and models, including Nvidia NIM microservices and Nvidia NeMo Retriever microservices on the Dataloop platform, gives customers the ability to speed up AI deployment and enhance their interactions with other tools and applications available on the platform.

The combination is claimed to deliver better data management and curation, as well as improved data automation workflows. The net outcome is assistance in advancing AI adoption and driving better business outcomes.

“Our collaboration with Nvidia marks a significant milestone in the AI development cycle,” Avi Yashar, co-founder and chief executive officer of Dataloop, said ahead of the announcement. “By combining Dataloop’s platform with Nvidia AI technologies, we are enhancing AI development and deployment, empowering our customers to leverage high-quality, optimized AI applications.”

The integration of Nvidia technology encompasses four elements, the first seeing Dataloop’s marketplace featuring access to the Nvidia API catalog, offering foundation models and application programming interface endpoints that allow users to incorporate leading AI models into data pipelines.

The ready-made AI solutions powered by Nvidia AI, such as pipelines for large language model applications, including summarization and chatbots, will be available for Dataloop customers. Nvidia NeMo text embedding-based data curation and search capabilities will be enabled within the indexed datasets managed on Dataloop.

The last element sees Dataloop receiving early access to Nvidia NIM and NeMo Retriever microservices, with the company integrating them into its platform. The integration will support text embeddings and LLM optimization. Combined, Dataloop aims to “revolutionize” how AI is deployed and utilized in the market, reducing cost and accelerating AI development.

Dataloop is a venture capital-backed startup, having raised $50 million in funding, according to Tracxn, including rounds of $11 million in 2020 and $33 million in 2022. Investors include F2 Venture Capital Inc., OurCrowd LLC, NextLeap Ventures LLC, Amiti Ventures LLC, Iconyc Inc. and SeedIL Ventures LLC.

Photo: Dataloop