Built to pass your security review

Quant research is sensitive by definition. Zerve runs inside your perimeter, on your infrastructure, under your controls.

Request demo

Three deployment models

Pick the one your risk appetite allows, not the one that suits us.

Air-gapped on premises

Runs on your hardware with no network path in or out. Suitable for the most restrictive environments.

Self-hosted VPC

Deployed inside your own cloud account on AWS, Azure, or GCP. Your data never leaves your perimeter.

Managed cloud

An isolated tenancy we operate, for teams deliberately reducing the number of systems they run internally.

CapabilityCloudSelf-hosted VPCOn-Premises
Instant setup
Managed compute infrastructure
Data never leaves your environment
No outbound network egress
Air-gapped operation

It runs on infrastructure you already operate

Standard building blocks, deployed by your team into your cluster. Nothing bespoke to maintain.

Installed with Helm charts into your own cluster, with compute jobs scheduled in a separate namespace.

Access and collaboration

SSO and role-based permissions, set per project and per data connection. Sharing a result never means sharing everything upstream of it.

Guardrails and constraints

Approved data sources, permissible tool calls, and required output formats, set centrally and enforced on every agent the platform runs.

Agent skills

Your analytical frameworks, approved methods, and research conventions, version-controlled and peer-reviewable rather than typed into a prompt.

Audit and attribution

Every data query, model call, transformation, and output is logged with attribution, and the full trace is inspectable and exportable.

Deterministic output

Agents leave behind editable code rather than a transcript. Any step can be rerun independently and produce the same result.

Models under your control

Run frontier models on your own keys, your internal gateway, or fully local with no external calls.