Human lead.
Agent crew.

CentaurQA tests software and builds AI automation with QA engineering discipline. Agents do the repetitive work; experienced people stay accountable for the decisions.

The experience behind the crew

Senior quality engineering, built into every agent.

More than ten years of test automation and enterprise software, now combined with hands-on experience building agentic AI systems.

10+years in QA & automation
10engineers led in one team
5industries
ISTQBAdvanced Test Analyst

What the human lead does

  • Sets the test strategy and the risk priorities
  • Designs automation architectures from scratch
  • Reviews and signs off the work at every gate
  • Builds CI/CD quality pipelines
  • Designs agent orchestration, evaluation and guardrails
  • Mentors your team and hands the process over

What the agent crew does

  • Analyses requirements and finds documentation defects
  • Designs tests on the right layer
  • Automates them in your stack and opens pull requests
  • Runs, stabilises and classifies failures
  • Selects risk-based regression for each release
  • Reports results to Jira/Xray with full traceability

Industries

HealthcareAutomotiveCybersecurityEducationOnline gaming

Stack

PythonJavaPlaywrightSeleniumRobot FrameworkREST AssuredrequestsREST · GraphQLKafkaMongoDBSQLAWS Lambda · S3 · SecretsDockerJenkinsGitHub ActionsGitLab CIJira · XrayMulti-agent systems · MCP
Security & trust

How we work with your code.

Your infrastructure first

Our default is to run the agents in your CI or on your machines, with your own AI provider account. We agree the setup with you before any work starts.

Least privilege by default

We ask for read-only access where possible. Merges, protected branches and bug reports stay behind human approval.

Traceable decisions

Each gate is designed to leave a record of who approved what, when, and based on which report.

One company, one point of contact.

Tell us what you need. We start with a free 30-minute call, with no sales pitch.