Welcome to Kubiya

Kubiya provides you with everything you need to get reliable, determinstic AI automation for the enteprise

🚀 Why Kubiya?

AI-Native Platform

Generate complex workflows using natural language. Our ADK provider uses cutting-edge LLMs to transform your requirements into production-ready automation.

Universal Integration

Connect anything: containers, APIs, databases, cloud services, AI models. If it has an interface, Kubiya can orchestrate it.

Deterministic Execution

Unlike agent frameworks that wander, Kubiya provides rails for AI. Every workflow step runs in isolated containers with defined boundaries.

💡 Key Features

🤖 AI Workflow Generation

from kubiya_workflow_sdk.providers import get_provider

# Generate workflows with AI
adk = get_provider("adk")
result = await adk.compose(
    task="Create a zero-downtime deployment pipeline with health checks",
    mode="act"  # Instantly execute!
)

from kubiya_workflow_sdk.providers import get_provider

# Generate workflows with AI
adk = get_provider("adk")
result = await adk.compose(
    task="Create a zero-downtime deployment pipeline with health checks",
    mode="act"  # Instantly execute!
)

from kubiya_workflow_sdk import workflow, step

@workflow
def deploy_app():
    # Every step runs in its own container
    build = step("build").docker(
        image="docker:latest",
        command="docker build -t myapp:latest ."
    )
    
    test = step("test").docker(
        image="myapp:latest",
        command="pytest tests/"
    ).depends("build")
    
    deploy = step("deploy").docker(
        image="kubectl:latest",
        command="kubectl apply -f k8s/"
    ).depends("test")

@workflow
def smart_incident_response():
    # Embed AI decision-making in workflows
    analyze = step("analyze").inline_agent(
        message="Analyze these system metrics: ${METRICS}",
        agent_name="sre-expert",
        ai_instructions="You are an expert SRE. Identify issues and recommend fixes.",
        runners=["kubiya-hosted"],
        llm_model="gpt-4o"
    )
    
    # AI generates remediation scripts
    fix = step("auto-fix").inline_agent(
        message="Generate safe remediation for: ${ANALYSIS}",
        agent_name="remediation-expert",
        runners=["kubiya-hosted"]
    ).depends("analyze")

🎯 Common Use Cases

CI/CD Pipelines

Build sophisticated deployment pipelines with canary releases, rollbacks, and health checks

Infrastructure Automation

Provision and manage cloud resources across AWS, GCP, Azure with unified workflows

Incident Response

Automate runbooks with AI-powered analysis and remediation

Data Processing

ETL pipelines that span multiple data sources and processing engines

Security Automation

Compliance checks, vulnerability scanning, and automated patching

ChatOps Integration

Connect workflows to Slack, Teams, or AI assistants via MCP

📚 Quick Start

1. Install the SDK

   pip install kubiya-workflow-sdk
   

2. Set your API key

   export KUBIYA_API_KEY="your-api-key"
   

3. Create your first workflow

   from kubiya_workflow_sdk import workflow, step, run
   
   @workflow
   def hello_kubiya():
       step("greet").shell("echo 'Hello from Kubiya! 🚀'")
   
   # Run it!
   run(hello_kubiya)
   

🔗 Integration Ecosystem

Docker

Kubernetes

AWS

Google Cloud

Azure

GitHub

OpenAI

Any Database

🛠️ How It Works

📖 Learn More

Quick Start Guide

Get up and running in 5 minutes

Core Concepts

Understand Kubiya’s architecture

Workflow Examples

Explore real-world use cases

API Reference

Complete API documentation

🤝 Join the Community

DiscordGitHubTwitter

---

Ready to revolutionize your automation? Get your API key and start building intelligent workflows today!