The way of Scale AI to become a $7.3 billion company, it relied on real image, text, voice, and video data. Now, it's using that foundation to get into the synthetic data business, one of the most popular and emerging categories in AI.
They announced an early access program to synthetic scale, a product machine learning engineers can use to improve their existing real-world data sets, according to the company. Scale hired two executives to develop this new division of their business. Scale brought in Joel Kronander, who previously led machine learning at Nines and was a former imaging engineer at Apple working on 3D mapping, as its new head of synthetic data. The company has also hired Vivek Raju Muppalla as its director of synthetic services. Muppalla was previously director of AI and simulation engineering at Unity Technologies.
Synthetic data is what it sounds like: fake data created by machine learning algorithms instead of using real-world information. It is a powerful and useful tool for generating data, such as medical images, when privacy is a key issue. Developers can use this synthetic data to add more complexity to their learning models and help remove biases that can often be found in real-world data sets.
Scale started by combining data with real images, text, voice, and video tagged by the team to provide autonomous vehicle companies with the data needed to train their machine learning models and to develop and deploy robotaxis, autonomous trucks, and automated bots used in warehouses and warehouses. in- logistics deliveries. The startup has since transformed into a data management platform company with clients spanning government, financial, e-commerce, autonomous vehicle, and enterprise industries.
Founder and CEO Alexandr Wang described his new offering as a hybrid approach to data, akin to lab-grown meat.
“We start with real data, just like raw lab meat starts from real animal cells, and then we grow, iterate and create the product from there.” By using real data as a source to create synthetic data, the company can offer a truly unique and powerful offering for customers, Wang said.
Clients Scale They saw this opportunity. The company's push into synthetic data was in response to demand from its customers, Wang told TechCrunch, noting that they started developing the product less than a year ago. Autonomous vehicle technology developer Kodiak Robotics, Tractable AI and the US Department of Defense have selected Scale for its new synthetic data product, Wang said.
Scale, which currently employs about 450 employees, sees synthetic data as a priority in 2022 and an area it will continue to invest in as it develops its product line. But that doesn't mean they will leave the real data business. Wang sees synthetic data as a complementary tool that will help developers “get more for their investment in algorithms and AI.
For example, autonomous vehicle companies often use simulation to recreate real-world scenarios and play them back to see how the autonomous system will use them. But real-world data may not provide the desired scenarios.
"You don't run into real-world situations too often where there could be, say, 100 cyclists crossing at once," Wang explained. "We can start with real-world data and then synthetically add all the cyclists or all the people." and then, that way, you can train the algorithm correctly.”
