Privacy Policy

Luceris LLC operates the Luceris platform. Luceris is sold to organizations, not to individual consumers. If you use Luceris as an employee or contractor of a customer organization, that organization controls your access and its agreement with us governs how its data is handled.

Two broad categories:

• Account and identity data — names, email addresses, organization names, roles, and memberships, used to authenticate you and determine what you can access.
• Operational data from your systems of record — work orders, comments, assets, properties, and regions synced from your CMMS or ERP. This can include personal details that live inside those records, such as requestor and assignee names, email addresses, phone numbers, service addresses, and free-text notes.

When you use the Luceris agent in Microsoft Teams, we also process your Teams identity and the content of the messages you send to the agent, so it can answer you and match you to your Luceris account by email address.

Operational data is pulled from your systems of record over HTTPS on a recurring incremental sync. Your source systems remain authoritative. Approved changes are written back to them through the same managed integrations, and every write is journaled.

You may also upload files directly. Those uploads go straight from your browser to private blob storage using short-lived, scoped credentials, so the file bytes never pass through our application servers or our background event stream.

All data is hosted with managed cloud providers in the United States. Application compute runs on Vercel. The primary system of record is a managed PostgreSQL database, with a replicated analytical copy used to serve agent queries, plus managed services for application state, caching, background job orchestration, and file storage. Your source system credentials are held in a dedicated secrets manager and referenced only by identifier.

Luceris uses third-party AI models for inference and reasoning only. Three commitments matter here:

• Your data is never used to train or fine-tune any model. Not ours, not a provider's. We operate no proprietary or fine-tuned models.
• Personal details are tokenized before prompts reach a model provider. Names, email addresses, phone numbers, addresses, and similar identifiers are replaced with stable placeholder tokens in our own infrastructure. Those tokens are resolved back to real values only in your browser, and only for users authorized to see them.
• Retention is restricted at the provider. We enforce zero data retention on our OpenAI traffic. Other providers are governed by their contractual terms.

Agents read your data through read-only views that coarsen sensitive columns, and cannot take action against your systems of record without explicit human approval.

We use the following categories of subprocessor, scoped to the features your organization has enabled: cloud hosting and compute, managed databases and analytical storage, application state and caching, background job orchestration, file storage, authentication and transactional email, AI model and embedding providers, document extraction (only where that feature is enabled), and application monitoring and tracing. Microsoft is a subprocessor only where the Microsoft Teams integration is in scope.

A current, named subprocessor list is available on request — write to support@luceris.ai.

Agent transcripts, generated outputs, and the placeholder-token maps described above are retained for the life of the workspace they belong to, and are deleted when that workspace is torn down or its agent state is reset.

Complete copies of the records we sync from your systems of record are retained for as long as your organization remains a customer, so that history stays inspectable over time. We do not currently apply a fixed time-based expiry to that source data. If your organization requires defined retention periods or a data subject access process, contact us and we will scope it with you as part of your agreement.

On termination, customer data is deleted on request. Backups and replicated copies are removed on their own cycle rather than instantaneously.

• Luceris is SOC 2 certified.
• All data is encrypted in transit using TLS 1.2+. We operate no unencrypted transport anywhere in the platform.
• Data is encrypted at rest across all managed data services.
• Every data-access path is scoped to a single organization, and that scoping is enforced mechanically in continuous integration rather than left to convention.
• Every request is bound to one organization by a verified token claim; access is deny-by-default.
• Role-based access control maps to your organizational structure, with an additional workspace-level boundary inside each organization.
• File processing runs in isolated sandboxes with all outbound network access denied.
• Every approved write back to your systems of record is logged with the authorizing user and full decision context.

Access to Luceris is administered by your own organization. To correct or delete your personal information, or to ask what we hold about you, contact your Luceris administrator or write to us directly and we will route the request. In Microsoft Teams you can turn agent notifications off at any time by sending the agent the message notifications.

We will update this page when our practices change, and will revise the date above when we do. Material changes affecting customer organizations are communicated through the account relationship rather than by page edit alone.

Questions about this policy, our subprocessors, or our security posture: support@luceris.ai. For details on the controls summarized above, see our security overview.