Trust

Deploy where your data lives.

Your environment, your model, your engineers’ approval on every change. A verified answer to the questions your security, risk and procurement teams will ask.

Request full reports, policies and questionnaires
Deployment · your accountLive
Data plane
Your Kubernetes cluster, your region
Identity
Your IAM roles and network policies
Model
Your approved endpoint
Changes
Approved by your engineers, kept in Git
Runtime
Flink · Kafka · Postgres · Iceberg
Open-source core, Apache 2.0.

01 / Deployment

Three ways to run it. You choose.

  • In your cloud account or on-premises

    Full isolation and full control. The data plane runs in your Kubernetes cluster under your IAM.

    Default for regulated workloads

  • Dedicated environment

    A single-tenant environment operated by DataSQRL for your organisation.

    Managed, isolated

  • Multi-tenant

    Shared infrastructure operated by DataSQRL at the lowest cost.

    Non-regulated workloads

Pipelines reach your databases, Kafka and object storage over private networking. The only public endpoints are the APIs you choose to build and expose. In the managed options, every customer is isolated in its own namespace and security group.

02 / AI and your data

Complete control and transparency.

  • Your model, your endpoint

    In your environment, the agent calls the model endpoint you configure, such as your own Amazon Bedrock, Azure OpenAI or Google Vertex AI deployment, or another provider your organisation has approved.

  • Nothing ships unverified

    Nothing an agent writes reaches production without compilation, validation, simulation and an engineer’s approval.

  • Not used for training

    Your data, schemas and code are never used to train models.

  • Everything is recorded

    Specifications, test results, approvals and deployments are kept in Git and in each pipeline’s lineage, so every change can be explained after the fact.

03 / Security and compliance

Audited, encrypted, and under your controls.

  • SOC 2 Type II: our controls are tested by an independent auditor. The report is available on request.
  • Runs in your Kubernetes cluster, under your IAM and network policies.
  • Background checks and security training for every employee.
  • Incidents and service status are posted on status.datasqrl.com.
  • APIs authorised with JWT-scoped rules and parameterised queries.
  • Open source: your security team can review the code that touches your data.
AICPA SOC 2 Type II

Information security policy (PDF) →

04 / Governance

Evidence on every change, not on request.

  • Lineage, schemas and a full data-flow graph on every compile.
  • Validators for PII, residency, retention and naming, enforced before deployment.
  • Every change approved as a pull request and kept in Git history with its test results.
  • Monitoring, alerting and history for every pipeline in production.

05 / Continuity

No lock-in, by design.

The core is Apache 2.0 on standard engines: Flink, Kafka, Iceberg and Postgres. The generated deployment assets are yours and run on those engines without DataSQRL. No proprietary runtime, no data format lock-in.

06 / For your review

Documents for your security and procurement review.

Our SOC 2 Type II report, security questionnaire, policies and architecture details are shared on request.

Request full reports, policies and questionnaires

07 / First step

Start with one use case. Decide on evidence.

A scoped Proof of Value on a use case you already have, with success criteria agreed before we start.

See results in weeks How the Proof of Value works →