On this page
Start hereIntroduction
Introduction
qbrix is an adaptive optimization engine delivered as a cloud service. You give it a set of variants and a way to score outcomes; it serves the best-performing variant on every request and keeps getting better from what happens next.
It runs selection and learning on separate paths, so deciding is fast no matter how much feedback you send. See How qbrix works for the timings that follow from that.
Key concepts
| Concept | Description |
|---|---|
| Pool | A named collection of arms (variants). For example, a pool of homepage hero images. |
| Arm | A single variant within a pool. Each arm has an index, a name, and optional metadata. |
| Experiment | Links a pool to a policy (algorithm). Defines how arms are selected and trained. |
| Policy | The algorithm qbrix uses to decide which variant to serve — and how it learns from each outcome. Available across stochastic, contextual, and adversarial categories, plus an auto mode that runs a portfolio of learners and adaptively routes traffic to the best performer. |
| Feature Gate | Controls experiment traffic with rollout percentages, scheduling, and targeting rules. |
| Feedback | A reward signal sent after selection. It's what the policy learns from. |
| Context | Optional request-level information (user features, metadata) used by contextual policies. |
How you'll use it
- Sign up at cloud.qbrix.io and create an API key
- Create a pool with arms via the API or console
- Create an experiment linking the pool to a policy
- Integrate select and feedback calls in your application
- Monitor and optimize from the console dashboard
What's next
- Quickstart — create your first experiment in minutes
- How qbrix works — the two paths, the timings, and the guarantees
- Pools & experiments — data model and lifecycle
- Feedback & rewards — the learning loop
- Policies — the algorithms behind every decision, and how to choose one
- API reference — full HTTP endpoint documentation