Snorkel AI empowers Fortune 500 enterprises, government agencies, and other AI innovators to solve their most impactful business problems with a transformative approach to data-centric AI. The company's flagship AI development platform Snorkel Flow is powered by state-of-the-art techniques in programmatic data labeling, weak supervision, and - for some use cases - foundation models/large language models.

Snorkel AI empowers Microsoft Azure customers to scale automation to the “iceberg under the surface” of valuable but unstructured enterprise data. Our unique approach equips data science and machine learning teams to adapt ML models of all sizes including multi-billion parameter large language models using data-centric AI workflows and programmatic labeling,acceleratinge AI application development 10-100x.

Snorkel Flow integrates seamlessly with Azure AI Services, enabling data science teams to quickly generate high-quality training sets over complex, highly variable data. Highlights include:

Data stored in Azure Blob Storage, Azure Data Lake Storage (ADLS), and other repositories can be easily consumed by Snorkel Flow.
If needed, Snorkel Flow can orchestrate document content preprocessing including OCR, layout information, and more through Azure Form Recognizer.
Data can then be labeled programmatically using a data-centric AI workflow in Snorkel Flow to quickly generate high-quality training sets over complex, highly variable data. Snorkel Flow includes templates to classify and extract information from unstructured text, native PDFs, richly formatted documents, HTML data, conversational text, and more.
Newly labeled datasets can be used to train standalone custom Azure Form Recognizer models.
Performance analyses can be auto-generated over custom Azure Form Recognizer models to guide next steps.
Models and results can be uploaded directly to Azure Blog Storage.
Snorkel Flow is Kubernetes-native and runs on AKS.

Snorkel Flow

by Snorkel AI