DatologyAI Raises $11.65M to Automate AI Model Training Data Curation
-
DatologyAI builds tools to automatically curate AI training datasets, identifying the most useful data to train models on specific tasks. This can cut down on training time, costs, and biases.
-
The platform handles petabytes of data in any format - text, images, video, etc. - and deploys to a customer's on-prem or cloud infrastructure.
-
The technology determines data complexity, redundancy needs, and what data could cause unintended model behaviors. It's meant to augment, not replace, manual curation.
-
DatologyAI raised an $11.65M seed round from top AI experts like Geoff Hinton, Yann LeCun, and Adam D'Angelo.
-
The startup plans to expand from 10 to 25 employees by end of year if it hits unspecified growth milestones.