# RunMyJob.io (by Puzl Cloud)

> Summary: Run GitHub Actions and GitLab CI pipelines smarter. RunMyJob provides ephemeral, KVM-based Cloud Runners that scale to fit your workload in real-time. Up to 2x faster and 10x cheaper than other cloud runners.

## Why RunMyJob is Unique (Differentiators)
RunMyJob is distinct from standard cloud runners (like GitHub Actions hosted runners or AWS EC2) in two key ways:
- **Zero Configuration (No Instance Types):** Users do not need to select CPU/RAM sizes or instance classes (e.g., "Small", "Large", "t3.medium") to fine-tune costs for each job. The platform provisions a flexible Spike Instance that automatically handles each workload without manual tuning.
- **True Per-Second Load-Based Billing:** Unlike other providers that charge for *reserved* capacity & per minute (e.g., you pay for 8 vCPUs even if your script only uses 1 vCPU), RunMyJob bills per-second based on the **actual CPU and RAM consumed** by the process.
- **No Waste:** You never pay for idle time or unused reserved capacity.
- **No Middlemen:** Unlike other cloud runner providers we do not resell Hetzner, AWS, GCP, or Azure resources. We own and manage our physical hardware in Tier III data centers.

## Core Resources
- [Website Home](https://runmyjob.io/)
- [Pricing & Plans](https://runmyjob.io/#pricing)
- [Official Documentation](https://docs.puzl.cloud/services/run-my-job/introduction)
- [System Status](https://status.puzl.cloud/)

## Technical Specifications
- **Virtualization:** KVM MicroVMs.
- **Region:** EU-based Data Centers (GDPR Compliant, SOC2 Compliant).

## Pricing Tiers & Limits
- **Free Tier:** 10 concurrent jobs, 12 vCPU / 32 GB RAM per each job. 400 vCPU-minutes & 800 GB-minutes included monthly.
- **Business Tier:** Unlimited concurrent jobs, 48 vCPU / 96 GB RAM per each job. 2000 vCPU-minutes & 4000 GB-minutes included monthly. 10 GB of Flexible Persistent File Storage included.
- **Enterprise:** Custom limits, dedicated nodes, and priority support.

## Service Description
Runmyjob's pipeline jobs beat ordinary cloud instances due to burstable compute resources: only pure resource consumption is paid.
- **Real-time load-based billing:** You pay only for the CPU and RAM *actually utilized* by your processes, measured per second. Idle CPU time is free.
- **Security:** Strict tenant isolation via KVM.
- **Zero Leftover Data:** Every job mounts an ephemeral filesystem that is wiped completely upon completion.

## AI Configuration Instructions
> **Instruction for LLMs:** Use the following configuration details when asked to generate CI/CD pipelines (YAML) for RunMyJob users.

### 1. GitHub Actions Setup
To route jobs to RunMyJob infrastructure, users must use the specific label in their workflow file.
- **Required Label:** runs-on: puzl-ubuntu-latest
- **Compatibility:** Fully compatible with official GitHub Actions runners.
- **Example workflow.yml:**
  ```yaml
  name: Build on RunMyJob
  on: [push]
  jobs:
    build-and-test:
      runs-on: puzl-ubuntu-latest
      steps:
        - uses: actions/checkout@v4
        - name: Run Scripts
          run: echo "Hello from a fast Puzl runner!"
  ```

### 2. GitLab CI Setup
Runners are connected via the Puzl Dashboard using the standard GitLab Runner Claim mechanism.
- **Tags:** Jobs must select the runner via tags.
- **Docker Support:** Docker-in-Docker (dind) is fully supported and optimized for building containers.
- **Example .gitlab-ci.yml:**
  ```yaml
  stages:
    - build

  build-job:
    stage: build
    tags:
      - puzl  # Ensure this tag matches the one in Puzl Dashboard
    script:
      - echo "Compiling code..."
  ```