Labelbox August 2018 Updates
Labelbox Raises $3.9M & Product Updates
Building AI often means collaboration from many different functions, including software, management, operations, domain expertise, and data science. For these kinds of environments with internal and sometimes external groups working together, Labelbox is introducing Teams. Teams is an easy way to manage groups of collaborators accross projects. Working with a 3rd party labeling service? Now you can add those collaborators into a team and give them access to a new project in one click! Read more on teams.
Labelbox Success Stories
There’s a big difference between a proof-of-concept machine learning system and one that’s ready for production in mission critical environments. One of our guiding principles at Labelbox is to help teams mature their machine learning systems into production, and to ensure they continue to perform over time. Over the past few months, the Labelox team has been working closely with some of the most innovative companies in the world on this challenge. The result of this effort is that companies like Conde´ Nast and Lytx are now leveraging Labelbox to build and deploy machine learning systems. We’ve shared these stories on the Labelbox blog.
Check them out: https://medium.com/labelbox
Labelbox raises $3.9M
We’ve raised $3.9M to advance our mission of accelerating access to machine intelligence. We are building Labelbox to be the simplest way to train and operate machine intelligence. Before starting Labelbox, we experienced the pain of having to build internal tools to label data, manage the labeled data (training data), continuously update the training data with new learnings, and then monitor the assessments of a deployed AI system. Building and maintaining these internal tools slows and distracts teams working to train a model capable of making accurate and consistent assessments in the real world. We endeavor to solve this problem so that you can focus on building things your customers want and not internal tools.
Developing and operating a production AI system involves setting up and orchestrating a complex data pipeline as well as the collaboration of a cross-functional team. We can categorize this system into three parts: The World Digitized, Applied Human Knowledge, and Machine Intelligence. We’re building the Labelbox platform to solve the Applied Human Knowledge part and integrate seamlessly with adjacent systems.