EDN is currently in early evaluation deployment. The practices below reflect our current architecture as we prepare for broader access.
Hosted on Microsoft Azure, Australia East region.
Multi-tenant middleware. Data is scoped per tenant at both the application and database layer.
Per-user, per-tier rate limiting enforced at the middleware layer.
If you believe you have found a vulnerability in EDN or the TensorPRO infrastructure, please contact us directly.
TensorPRO uses the following third-party subprocessors to operate the EDN Memory Engine. We update this list when we engage a new subprocessor, and notify customers with an active agreement before a new subprocessor starts processing personal data.
The large language model used with EDN is chosen and connected by you. Your model provider is your own processor, not a TensorPRO subprocessor.
We do not use the content you submit to train models for other customers or to build competing products. For more on how we treat your data, see our Responsible AI page.
For a Data Processing Agreement, security documentation, or a security questionnaire, contact contact@tensorpro.ai.
Memory infrastructure for AI systems. No fine-tuning. No model changes.
Integration guides, API reference, and architecture documentation for developers.
View documentationBenchmark results, methodology, and the provenance model behind EDN Memory Engine.
View research