New Release! Git-backed Machine Learning Model Registry for all your model management needs.
Security and Privacy at ReciprocateX
ReciprocateX's products are trusted by thousands of organizations. We place security as a core requirement during product development and maintain regular internal and external security assessments. Customers may request all relevant security documents from us as they evaluate our products.
Data security and privacy
ReciprocateX products collect and use only necessary data to function properly. We retain customer data for as long as an account is active, as needed to provide services to customers, or in accordance with the agreement(s) between ReciprocateX and the customer, unless ReciprocateX is required by law to dispose of it earlier or keep it longer. ReciprocateX does not use any personal information collected in the course of doing business for commercial purposes. Please read our full privacy policy at https://dvc.org/doc/user-guide/privacy.
Open-source
Most of our MLOps solutions are open source and thereby subject to public review. Security related to our open source tools would be managed by the user as our tools are downloaded locally. Users manage their own credentials and security policies across resources like clouds, storage, and Git service. There are logging functionality that send anonymized usage data back to ReciprocateX. Users may opt out of this logging. We'll promptly address any security issues that are brought up by the community. Please let us know at https://github.com/ReciprocateX.
Compliance
ReciprocateX is SOC2 compliant! See the report here.
You can read the key learnings from our compliance experience here.
Studio permissions and security
As part of our GitOps philosophy, Studio only takes as much information as necessary from your Git service to display experiments, data sets used, metrics, and hyper parameters. Studio only has access to repositories that customer Git services allow. By default, Studio does not access any of the actual data used across your models. Your data remains protected by your cloud credentials (e.g., AWS login, etc.). You may allow Studio access to storage for additional information to be displayed by Studio, but this is optional. Access controls to repositories may be granularly managed directly through a customer's respective Git service (GitHub app, GitLab admin settings, etc.).
Bug Bounty Program
To maintain and improve the security of our systems and tools, we are happy to work together with the security community. We believe in a responsible collaborative model around vulnerability testing and reporting that will protect and benefit our customers and users. Please read our full Bug Bounty Program.
