Sama, the leader in providing data annotation and model validation solutions, has announced the launch of a multi-cloud integration strategy within its platform, offering customers both significant cost savings and easier compliance with both internal and internationally recognized security standards. Customers can now keep their data on one of the three major global cloud storage providers (AWS, Google Cloud and Microsoft Azure) and seamlessly and securely give Sama access to that data. With these new integrations, customer onboarding can be reduced to just one day, speeding the process up as much as 7x.
This faster onboarding process reduces the need for custom coding solutions to grant Sama access to customers’ servers, saving resources in addition to time. Sama now offers a complete suite of integrations including compatibility with Python Software Development Kits (SDKs) and the Databricks platform along with Okta, SAML and more. Additionally, Sama’s custom data pipeline integration with these technologies allows the company to work seamlessly with already-designed data flows, eliminating the need for client engineers to set up annotation and validation workflows and further reducing resource use.
To adequately address the many different edge cases that can lead to model failure in fields like AgTech, autonomous driving or ecommerce, an immense dataset is required. Sama’s new compatibility with major cloud providers allows us to access these datasets on behalf of our clients quickly, securely and cost-effectively. The faster we can complete onboarding, the faster we can assist our clients in getting their models to market.Duncan Curtis, SVP of Product and Technology, Sama
For annotation and validation projects with hundreds or even thousands of terabytes of data, as with 3D sensor fusion projects, the costs to transfer data from one cloud to another can quickly add up. With multi-cloud integration, Sama can save projects as much as $400,000 on up to 2,000 terabytes of data by transferring files from the customer’s cloud server to a server of the same architecture and in the same region.
Preventing security breaches remains a critical concern, due to the use of large amounts of data that potentially involve personally identifiable information. Sama’s multi-cloud integration ensures the company does not retain a copy of private data. Instead, Sama can securely access a customer’s files, make any modifications needed (such as filtering and generating 3D ground points), and then write the updated file back directly to the customer’s server. Files are always hosted by a customer’s server, ensuring they meet IT requirements and increasing data security. Sama’s Platform 2.0 supports a high level of data security through enterprise-grade SSO integrations, GDPR and CCPA compliance, and ISO and TISAX certifications.
Sama also offers integrations with a Python SDK built on top of Sama’s existing APIs and Databricks, ensuring less development time for onboarding or custom scripts to automate the sending of data. In addition, clients can send Sama data directly from the user-friendly Databricks platform, and Sama can then seamlessly transfer data back into the same data pipeline.
These newest integrations support Sama’s continued efforts to accelerate time to market and reduce total costs of model ownership. Leveraging its Platform 2.0 and fully-trained expert workforce, Sama has previously improved the speed at which a model could be brought to market by 3-4x by eliminating lengthy delays caused by rework. Platform 2.0 has successfully achieved a 99% client acceptance rate for AI training data through SamaAssure™, the industry’s highest quality guarantee, with an annotation delivery rate of 300+ million frames, 850+ million shapes and 10 billion annotation points a month.