US-based machine learning technology developer Snorkel AI publicly launched yesterday with $15m from investors including GV, an early-stage investment arm of internet and technology group Alphabet.

The round was also backed by venture capital firm Greylock Partners and In-Q-Tel, the strategic investment affiliate of the US intelligence community, as well as undisclosed additional investors.

Snorkel’s software automatically labels large datasets for training artificial intelligence (AI) algorithms, a task that can take months to perform manually.

Data labelling is required as algorithms observe tagged examples of the job they are to execute before running autonomously, but that task becomes harder when complex or sensitive information is involved.

The startup applies heuristic rules and functions to determine labelling parameters without a human supervising the process. Its clients include unnamed banks, government agencies and enterprise businesses.

The original version of this article appeared on our sister site, Global University Venturing.

Snorkel AI, a US-based machine learning modelling technology spinout of Stanford University, emerged from stealth yesterday with $15m from investors including GV, an early-stage corporate venturing arm of internet and technology group Alphabet.
The round was also backed by In-Q-Tel, the strategic investment affiliate of the US intelligence community, venture capital firm Greylock Partners and undisclosed investors.
Snorkel AI’s platform automatically labels large datasets for training artificial intelligence (AI) algorithms, a task that can take months to perform manually.
Data labelling is needed as algorithms observe tagged examples of the job they are to execute before running autonomously, but this becomes harder where complex or sensitive information is involved.
Snorkel applies heuristic rules and functions to determine labelling parameters without a human supervising the process. Its clients include unnamed US banks, government agencies and enterprises.
The spinout emerged from research at Stanford’s AI Lab by investigators including Chris Ré, associate professor in the university’s Department of Computer Science, along with his former PhD understudies Paroma Varma,  Braden Hancock, Henry Ehrenberg and Alex Ratner.
Ratner has since joined University of Washington as an assistant professor and also acts as CEO of Snorkel AI. Ehrenberg moved onto social media group Facebook in a machine learning role after completing his Stanford research in 2017.
AJ Bertone, an investment partner at In-Q-Tel, said: “The time, expertise, and costs involved in labelling training data present significant challenges to the US government in applying AI to missions of national security.
“Snorkel AI provides a revolutionary capability that can greatly reduce the level of effort required to develop mission-ready machine learning models by addressing this critical data problem.”