Vedric has three parts: an endpoint agent, a cloud service, and an operator console. They are connected by signed messages and tenant-isolated at every layer. Content collection is limited to one documented exception, a bounded and redacted PowerShell script preview. Everything executed is logged. Nothing written can be quietly undone.
Endpoint
Runs as a managed service under OS-sealed credentials. Collects primarily behavioral metadata from the host, identity, and network layers, plus a documented bounded PowerShell script preview where host logging is enabled.
Cloud
Baselines behavior per user and per peer group, correlates deviations into storylines, and runs triage against a strict budget. Every query is tenant-scoped. If the AI layer is unavailable, baseline detection keeps running.
Operator
Analysts can correlate related activity into storylines. Remote response commands are admin-gated, reason-required, signed, and audited, but remain default-off pending deployment-specific safety drills.
Endpoint behavior is collected across several surfaces by default. Deeper visibility modes can be enabled per tenant. What is never collected is documented and refused in code.
User identity is anonymized on the endpoint. The backend stores a per-machine one-way token, not the raw identity. Your team can label that token after an independent local investigation; Vedric cannot reverse it.
Two layers run in parallel. Behavioral baselines (per user, per peer cohort, per endpoint) surface drift from the user's own pattern. Rule-based detection catches well-known-bad patterns from day one. A warm-up period prevents the baseline layer from generating noise while it settles; the rule layer is active immediately.
Deviations and rule hits on the same endpoint and user within a short window become a single correlated storyline.
Triage runs inside a fail-closed budget and circuit. It reads tenant-scoped data only and never reaches the endpoint. If it is unavailable for any reason, baseline detection continues.
Remote response is a default-off preview. Its command paths are admin-gated, reason-required, signed, and audited, but they remain unavailable until the deployment passes endpoint safety and recovery drills.
Every decision lands in an append-only audit. Edits and deletes are refused at the storage layer. The forensic trail is preserved without relying on operator discipline.
The refusal list lives in the agent source. PowerShell script previews are the one bounded exception: they require host logging, are truncated and scrubbed, and machine-policy management is opt-in.
Every tenant-scoped query is automatically filtered. The guarantee is tested on every build: if a new table would create a cross-tenant path, the build fails.
Agent binaries are signed with a key that never lives on the server. A cloud compromise alone cannot produce a valid update; the attacker would need offline key custody too.
The application database role is refused edits and deletes on audit rows. Customers with a database-owner or infrastructure-admin threat model should also retain exports in an independently controlled system.
Triage runs behind a fail-closed control layer: if the check that decides whether AI may be used cannot be completed, it resolves to no. An admin can disable AI triage for their whole organization, and an operator kill-switch propagates across the platform quickly. Baseline behavioral detection continues regardless. Detection does not depend on the AI layer.
Remote commands require an admin role and written reason and use a signed channel, but remain disabled until deployment-specific endpoint and recovery drills are complete.
Quoted per deployment after a short scoping call. The quote depends on endpoint count, deployment option and support.
SaaS
Vedric's managed cloud, logically isolated per tenant. Available now.
Deployment options
Three options for where your detection data is stored. The pricing page lists what changes with each.
Early access
Direct-to-engineering support while the design-partner group is open.
More on specific detections: insider risk detection, data exfiltration detection, privileged user monitoring and insider threat use cases.
A 30-minute call with an engineer who builds Vedric. No slides.