Last Updated: February 23, 2026
This Privacy Policy ("Privacy Policy") describes how Nanite Foundry, LLC ("Company," "we," "us," or "our") collects, uses, discloses, retains, and otherwise processes information in connection with its services, software, APIs, proxy and routing infrastructure, memory optimization systems, dashboards, and related functionality (collectively, the "Service").
This Privacy Policy applies to information processed when you access or use the Service. Capitalized terms not defined in this Privacy Policy have the meanings set forth in the Terms of Use.
The Company operates a technical infrastructure and optimization service that processes user-provided inputs for the purpose of routing, summarization, memory handling, efficiency optimization, and transmission to third-party artificial intelligence providers designated by users.
The Company does not operate or control third-party AI models and does not determine the substantive content of AI-generated outputs.
Depending on the context, the Company may act as a data controller or a data processor. Where the Company processes information on your behalf in connection with your use of third-party AI providers, you remain responsible for determining the lawful basis for such processing and for compliance with applicable data protection laws.
We collect information you voluntarily submit through the Service, including:
When you access or use the Service, we may automatically collect:
We may receive limited information from third-party service providers, including infrastructure providers, payment processors, and AI platforms, in connection with operating the Service.
In connection with Observability Services, we may collect and process:
System-generated analytical outputs derived from Customer Data may be retained by the Company and used in aggregated or anonymized form for product improvement and benchmarking.
We use information for the following purposes:
To provide, maintain, secure, and operate the Service, including processing, routing, summarizing, caching, and transmitting information to third-party AI providers designated by you.
We use information to:
Except as stated in Section 3.4, we do not use Conversation Data to train or fine-tune general-purpose machine learning or large language models.
To comply with legal obligations, enforce our Terms of Use, prevent misuse or abuse of the Service, and protect the rights, safety, and property of the Company and others.
Company will not use, and will not permit any third party to use, your information, including without limitation any conversation data, prompts, responses, inputs, outputs, metadata, or other information submitted to or generated through the Services ("Conversation Data"), to train general-purpose large language models or foundation models.
The Company does not sell Conversation Data and does not use such data for targeted advertising.
We may reprocess, replay, simulate, or re-evaluate historical prompts, conversation data, and instruction files to generate regression analyses and performance reports.
Such reprocessing may involve transmitting data to third-party AI providers designated by you.
When the Service is used solely as a proxy, routing, or optimization layer and Observability or Backtesting functionality is not enabled, we do not retain complete conversation transcripts after transmission to the third-party AI provider designated by you.
In such cases, Conversation Data is processed transiently for routing, transformation, summarization, or optimization purposes and is not persistently stored by us, except for limited system logs, security monitoring records, and infrastructure diagnostics consistent with ordinary operation of technical services.
When Observability Services or Backtesting functionality are enabled by you, we may temporarily store complete Conversation Data for the purpose of:
Complete Conversation Data retained for these purposes will not be stored in active systems for longer than seven (7) days from initial processing, after which it will be deleted, except as required for security logging, fraud prevention, legal compliance, dispute resolution, or backup integrity.
When Memory functionality is enabled by you, we may extract, derive, and store structured factual elements from Conversation Data ("Memory Data"), including preferences, configuration data, identifiers, or other contextual facts determined by automated systems.
Memory Data is stored solely for the purpose of making such facts accessible to your bot or workflow to enhance continuity and contextual performance.
Memory Data may include limited excerpts of Conversation Data necessary to preserve contextual accuracy. However, unless Observability or Backtesting functionality is separately enabled, we do not retain complete conversation transcripts in connection with Memory functionality.
Memory Data is retained for as long as the Memory feature remains enabled. Upon deactivation of the Memory feature, Memory Data will be deleted from active systems within thirty (30) days, except as required for legal compliance, fraud prevention, dispute resolution, or backup integrity.
We may process and retain incomplete or partial Conversation Data, including individual message snippets, truncated segments, token counts, or automated summaries, solely for the purposes of:
Such processing does not involve the storage of complete conversation transcripts outside the limited retention periods described above.
We do not use Conversation Data to train, fine-tune, or improve general-purpose large language models or foundation models. Conversation Data is used only as described in this Privacy Policy.
We may use information in aggregated, anonymized, or de-identified form to improve infrastructure reliability, routing logic, memory optimization, system diagnostics, performance benchmarking, abuse detection, and feature stability.
We do not use Conversation Data to train general-purpose large language models or foundation models.
Any internal system improvements are limited to infrastructure, reliability, security, and analytical feature enhancement and do not involve training generative AI models on user-submitted content.
We may share information in the following circumstances:
Information is transmitted to third-party AI providers designated by you, using credentials you supply, in order to generate AI outputs. Third-party AI providers may retain or process data in accordance with their own privacy policies and terms. The Company does not control or assume responsibility for the data handling practices of such providers.
We may share information with vendors and service providers who assist us in operating the Service, including hosting, infrastructure, analytics, and payment processing providers.
We may disclose information if required to do so by law, legal process, or governmental request, or if we believe disclosure is necessary to protect the rights, safety, or property of the Company or others.
Information may be disclosed in connection with a merger, acquisition, financing, reorganization, or sale of assets.
We retain information only for as long as reasonably necessary to provide the Service and fulfill the purposes described in this Privacy Policy.
Retention periods vary depending on feature usage:
Deletion from active systems does not guarantee immediate removal from backup systems, which are retained in accordance with standard disaster recovery practices.
We implement reasonable administrative, technical, and organizational measures designed to protect information processed through the Service. However, no system is perfectly secure, and we cannot guarantee absolute security.
Because prompts and conversation logs may contain sensitive information, users should avoid submitting confidential, regulated, or highly sensitive data unless necessary. We implement commercially reasonable administrative, technical, and organizational safeguards, but no system is entirely secure.
You acknowledge that you use the Service at your own risk.
You may stop using the Service at any time and may revoke third-party API keys at the source provider.
We will process deletion requests in accordance with applicable law. Certain information may be retained as required for legal compliance, dispute resolution, fraud prevention, backup integrity, or legitimate business purposes consistent with this Privacy Policy.
Depending on your location, you may have rights under applicable data protection laws, including rights of access, correction, or deletion. Requests may be subject to limitations and verification requirements.
Information may be processed and stored in the United States and other jurisdictions. By using the Service, you consent to the transfer of information to countries that may have different data protection laws than your jurisdiction.
The Service is not intended for use by children under the age of 13. We do not knowingly collect personal information from children.
We may update this Privacy Policy from time to time. The updated version will be effective as of the date stated above. Continued use of the Service constitutes acceptance of the updated Privacy Policy.