Quickstart Guide

Get DeployPilotOS running locally in under 3 minutes.

1. Product Overview & Business Outcomes

DeployPilotOS is a highly scalable, autonomous AI Site Reliability Engineer (SRE). By combining deterministic infrastructure management with OpenAI's advanced reasoning capabilities, DeployPilotOS aims to revolutionize enterprise DevOps and incident response.

🎯 Target ARR / MRR

Targeting $5M ARR within 18 months via Enterprise SaaS licensing ($10k/mo base platform fee + $500/node active monitoring fee). Expected $416k MRR by end of Q4.

💰 Cost Reduction

Reduces average MTTR (Mean Time To Resolution) from 45 minutes to under 2 minutes, saving enterprise clients an estimated $300,000 per hour of downtime prevented.

🚀 Market Position

Positioned to disrupt traditional incident management tools like PagerDuty by executing autonomous remediations rather than just triggering alerts.

2. Local Installation

Clone the repository and install dependencies (Node.js 18.x required):

git clone https://github.com/deploypilot/deploypilot.git cd deploypilot npm install cp .env.example .env.local # Add OPENAI_API_KEY="sk-..." to .env.local npm run dev

3. Cloud Deployment Strategies

DeployPilotOS is designed to be cloud-agnostic and container-native. Below are the standard operating procedures for deploying the Mission Control platform across major infrastructure providers.

AWS Elastic Kubernetes Service (EKS)

For AWS, we deploy DeployPilotOS using Helm charts integrated with AWS Secrets Manager.

  • Create an EKS cluster with managed node groups (t3.xlarge recommended for core services).
  • Provision an RDS PostgreSQL instance for persistent state and DynamoDB for high-throughput metric ingestion.
  • Deploy the deploypilot-agent daemonset across all worker nodes to collect low-level cgroup metrics.
aws eks update-kubeconfig --name dp-prod --region us-east-1
helm upgrade --install deploypilot ./charts/deploypilot -f values.aws.yaml

GCP Google Kubernetes Engine (GKE)

GCP deployments leverage GKE Autopilot for zero-ops cluster management and Cloud Spanner for global state synchronization.

  • Enable Workload Identity to securely bind K8s Service Accounts to GCP IAM roles.
  • Configure Pub/Sub topics to pipe raw logs directly into the DeployPilotOS Diagnosis Engine.
gcloud container clusters get-credentials dp-prod-gke --region us-central1
kubectl apply -k kustomize/overlays/gcp

Azure Kubernetes Service (AKS)

Azure deployments utilize Azure Key Vault for secret rotation and Azure Monitor integration.

  • Deploy using Terraform modules available in the /terraform/azure directory.
  • VNet integration is required to allow the DeployPilotOS agent to reach private AKS endpoint APIs safely.

OCI Oracle Cloud Infrastructure

For OCI, DeployPilotOS runs on OKE (Oracle Kubernetes Engine) utilizing ARM-based Ampere A1 compute instances for extreme cost efficiency.

  • Integrates with OCI Streaming Service (OSS) for Kafka-compatible event ingestion.
  • Requires configuring OCI Vault for secure OpenAI API key storage.