> ## Documentation Index
> Fetch the complete documentation index at: https://docs.liveops.adityas.net/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Runtime behavior patches for production services

# LiveOps

**Change production behavior without redeploying.**

Record traffic, simulate a scoped behavior patch, deploy it with canary controls, and roll back quickly when feature flags aren't enough.

## The Problem

When a downstream service degrades during an incident:

| Option            | Limitation                                    |
| ----------------- | --------------------------------------------- |
| **Redeploy**      | Too slow (30+ minutes)                        |
| **Feature flags** | Only toggle pre-compiled code paths           |
| **Service mesh**  | Can retry/timeout, can't synthesize responses |

None of these let you inject *new behavior* at runtime.

## The Solution

LiveOps separates **when** to intervene from **what** to do.

Deploy a behavior patch after replaying it against recorded request context. No redeploy. No pre-compiled variants.

## Key Properties

* **Safe by default** — Executes in under 10ms with hard timeout
* **Simulate before deploy** — Rehearse policies against recorded request context
* **Pure evaluation** — Deterministic and testable
* **Blast radius controls** — Canary %, tenant targeting
* **Incrementally adoptable** — Go middleware, CLI, and standalone core library

<Card title="Request Early Access" icon="rocket" href="https://calendly.com/saditya9211/30min">
  20-minute demo. No sales pitch.
</Card>