Snorkel Flow Update Offers Enhanced Enterprise Data Management Capabilities

Snorkel AI has announced a major update to its flagship data labeling, filtering, curation, and AI fine-tuning platform named Snorfel Flow. The latest update aims to address one of the most pressing challenges for companies looking to develop and deploy AI – integration of enterprise data with AI models. 

The Snorkel Flow update streamlines the integration of vast amounts of enterprise data into AI models. The platform can now be integrated directly with Google’s Gemini 3, Meta’s recently released Llama 3, and other models. This offers increased flexibility for businesses to choose the LLM best suited to their needs. 

The upgrade also features data source integration with Vertex AI, Databricks Unity Catalog, and Microsoft Azure Machine Learning to streamline access for data labeling. In addition, Snorkel Flow now supports programmatic labeling of multimodal data such as text, images, and audio. 

Snorkel Flow was launched in March 2022, enabling organizations to significantly accelerate AI application development and deployment with automated data labeling. The first version included features such as collaborative AI development and an Integrated ML modeling suite. 

The approach by Snorkel Flow for enterprise data management is to do programmatic labeling and iterative improvements to manage large volumes of data used for training AI models. Snorkel AI claims that the Snorkel Flow approach can reduce the time and cost of data labeling by 10-100x.  

The latest update builds upon the earlier version by offering a more streamlined workflow for managing the data labeling process. Users can now define labeling functions, manage data sources, and monitor label quality. These upgrades offer better utilization of resources to prepare enterprise data for AI training. 

“Enterprises are quickly hitting a wall with what they can achieve using off-the-shelf LLMs, and are seeing that the next wave of value will be unlocked by tuning LLMs on their unique data and use cases,” said Alex Ratner, co-founder and CEO, Snorkel AI. 

Ratner added, “As base LLMs become pervasive, including powerful open source options like Llama 3, the speed and accuracy with which data is continuously labeled and curated for fine-tuning and aligning LLMs becomes the key differentiator.”

Snorkel AI started as a research project in the Stanford AI Lab in 2015. In 2019, the startup launched from stealth mode announcing it had received $3 million dollars in seed money. By 2021, the research project had grown to secure several rounds of funding and was valued at a staggering $1 billion. The startup has partnered with some of the world’s largest companies including IBM, Apple, Intel, and Uber. 

The company specializes in data labeling,  data augmentation, and model training. The tools offered by Snorkel AI allow users to create high-quality training datasets more efficiently than traditional manual labeling methods.

As businesses continue to turn to AI at a rapid rate, more companies are offering training data services. Snorkel AI faces competition from large and small companies. Its top competitors in the data labeling market include CloudFactory, Labebox, and Scale AI. 

The upgrade to Snorkel Flow comes at a time when enterprises are looking to leverage AI across various data modalities. Whether it is structured or unstructured data, AI technologies are being applied to extract useful insights that can power business decision-making. With Snorkel Flow’s simplified data labeling and integration with powerful AI models, enterprises now have a tool that can unlock new possibilities for  AI technologies. 

Related Items 

Snorkel AI Partners with Microsoft to Unlock the Potential of Enterprise AI

Refuel Launches Platform to Automate Data Labeling Using LLMs, Having Already Delivered 10B+ Annotations Across Customers

Snorkel AI Accelerates Foundation Model Adoption with Data-centric AI