VentureBeat presents: AI Unleashed - An exclusive executive event for enterprise data leaders. Network and learn with industry peers. Learn More
Snorkel AI, a startup developing data labeling tools aimed at enterprises, today announced that it raised $35 million in a series B round led by Lightspeed Venture Partners. The funding marks the launch of the company’s Application Studio, a visual builder with templated solutions for common AI use cases based on best practices from academic institutions.
According to a 2020 Cognilytica report, 80% of AI development time is spent on manually gathering, organizing, and labeling the data that’s used to train machine learning models. Hand labeling is notoriously expensive and slow, with limited leeway for development teams to build, iterate, adapt, or audit apps. In a recent survey conducted by startup CloudFlower, data scientists said that they spend 60% of the time just organizing and cleaning data compared with 4% on refining algorithms.
Snorkel AI hopes to address this with tools that let customers create and manage training data, train models, and analyze and iterate AI systems. Founded by a team spun out of the Stanford AI Lab, Snorkel AI claims to offer the first AI app development platform, Snorkel Flow, that labels and manages machine learning training data programmatically.
An exclusive invite-only evening of insights and networking, designed for senior enterprise executives overseeing data stacks and strategies.
Application Studio will expand the Snorkel AI platform’s capabilities in a number of ways, the company says, by introducing prebuilt solution templates based on industry-specific use cases. Customers can leverage templates for contract intelligence, news analytics, and customer interaction routing as well as common AI tasks such as text and document classification, named entity recognition, and information extraction. Application Studio also provides packaged app-specific preprocessors, programmatic labeling templates, and high-performance open source models that can be trained with private data, in addition to collaborative workflows that decompose apps into modular parts.
Beyond this, Application Studio offers a feature that versions the entire development pipeline from datasets to user contributions. With a few lines of code, apps can be adapted to new data or goals. And they keep training data labeling and orchestration in-house, mitigating data breach and data bias risks.
Application Studio is in preview and will be generally available later this year within Snorkel Flow, Snorkel AI says.
Palo Alto, California-based Snorkel AI’s latest fundraising round brings the startup’s total raised to date to $50 million, which 40-employee Snorkel AI says will be used to scale its engineering team and acquire new customers. Previous investors Greylock, GV, In-Q-Tel, and Nepenthe Capital, along with new investor Walden and funds and accounts managed by BlackRock, also participated in the series B.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.