Analytics and AI that run where your data lives.
Query your databases, forecast demand, and run workflows inside your own environment. KEZEL checks each action against your policies and keeps a record of who accessed the results.
Deploy on-premises, in your cloud, or across both. Your team operates the agent beside your databases.
SOC 2 Type II attested, 2023–2025 · ISO 9001 certifiedrevenue by week, EU region, last quarter?
REVENUE · WEEKLY · EU
Choose a starting point.
Start with the job you need to do. Each component uses the same Zero Trust Orchestration runtime, so you can add others as your deployment grows.
Ask a question against the data you already hold.
Build dashboards, query in plain language, or forecast demand with KEZEL Insights Studio. Queries run beside your databases; readers access the results in your environment under your policies.
Explore analyticsExample: forecast inventory demand for the next four weeks.
- Dashboards and plain-language queries
- Forecasting and explained risk scores
- Column masks, lineage and access logs
The work runs beside your databases.
An agent executes the request in your environment. Results are consumed there too, through the policy gate. Open each step to see how a request runs.
01Define who can run and read
Your policy sets reader roles, row filters and column masks. The agent checks that policy when the work runs.
02Send an encrypted instruction
The control plane builds the instruction from schema metadata and intent. The agent pulls it over a connection it initiates.
03Execute beside the source
Queries, transformations and analytical models run in your environment. Computed results remain there.
04Read the result under policy
Dashboards and APIs consume results inside the same boundary. Each read records the reader and policy decision in the audit log.
Forecast demand or find the cases worth reviewing.
Explore forecasting and risk scoring with KEZEL Insights Studio. Choose a method to see a worked example. These use sample data; in a deployment, the models run against your own tables.
“How much?”
Estimate how spend affects weekly volume
Fit the relationship between campaign spend and volume using your own tables. The fitted line shows an estimate; an analyst still needs to check whether the relationship supports the decision.
regression · fitted relationship on sample data
campaign spend vs. weekly volume · fitted · sample data
See which project decisions need your attention.
KEZEL VIA keeps tasks, specs and agent run records in your deployment's store. Each member has a queue of approvals and open decisions. Select an example item to inspect it.
For team updates and blockers, KEZEL Check-in links each entry to the work it describes.
VIA-212 · Approval
The proposed job changes the KYB refresh schedule. A named approver must review the plan before this task can proceed.
Plan attached · risk threshold exceeded
Find a workload in your sector.
Explore deployment examples, from banking reviews to plant maintenance. See where the agent runs and how teams can use it.
Notes on deployment and data access.
Technical explanations of KEZEL's architecture and the review work it supports.
See KEZEL run a workload like yours.
Tell us which system holds your data and what you need to do with it. We’ll demonstrate on sample data, then scope a pilot with your security team.