← Back to Projects

LaunchApp Platform

GitLab CI/CD driven environment orchestration using Kubernetes, EKS, Helm, and ArgoCD

DevOpsKubernetesAWSInfrastructureCI/CD
View on GitHub

Overview

LaunchApp is a self-service platform that enables development teams to provision and manage their own environments using GitLab CI/CD, Kubernetes, and ArgoCD. The platform automates environment orchestration, making it easy for teams to deploy applications across multiple environments with consistent configuration.

Key Features

  • Self-Service Provisioning: Teams can create new environments through GitLab CI/CD pipelines
  • GitOps Workflow: All configuration managed through Git repositories
  • Multi-Environment Support: Dev, staging, and production environments
  • Automated Deployments: ArgoCD handles continuous deployment
  • Resource Management: Kubernetes namespaces and resource quotas per environment
  • Observability Integration: Built-in monitoring and logging

Tech Stack

  • Container Orchestration: Kubernetes (Amazon EKS)
  • GitOps: ArgoCD
  • Package Management: Helm
  • CI/CD: GitLab CI/CD
  • Infrastructure: AWS (EKS, VPC, ALB, Route53)
  • IaC: Terraform
  • Scripting: Bash, Python

Architecture

Platform Components

  1. GitLab CI/CD Pipelines: Orchestrate environment creation and updates
  2. Terraform Modules: Provision AWS infrastructure and EKS clusters
  3. Helm Charts: Package and deploy applications
  4. ArgoCD: Continuous deployment and synchronization
  5. Kubernetes Operators: Custom resource management

Environment Workflow

Developer → GitLab Merge Request → CI Pipeline → Terraform Apply → 
Helm Package → ArgoCD Sync → Kubernetes Deployment

Implementation Details

Environment Provisioning

  • Terraform modules create isolated namespaces
  • Resource quotas enforce limits per environment
  • Network policies ensure isolation
  • RBAC controls access permissions

Application Deployment

  • Helm charts define application configuration
  • ArgoCD monitors Git repositories for changes
  • Automated rollback on deployment failures
  • Health checks validate deployments

Monitoring & Observability

  • Datadog integration for metrics and logs
  • Prometheus for Kubernetes metrics
  • Custom dashboards per environment
  • Automated alerting for failures

Outcomes

  • Developer Productivity: Reduced environment setup time from days to hours
  • Consistency: Standardized deployment process across all teams
  • Reliability: 99.9% uptime for production workloads
  • Cost Optimization: Automated scaling reduced infrastructure costs by 30%
  • Self-Service: Teams manage their own environments without ops involvement

Technical Challenges

Challenge 1: Multi-Tenancy

Solution: Implemented namespace isolation with network policies and resource quotas

Challenge 2: Configuration Management

Solution: Centralized Helm values repository with environment-specific overrides

Challenge 3: Disaster Recovery

Solution: Automated backup and restore procedures using Velero

Best Practices Implemented

  • Infrastructure as Code for all resources
  • GitOps workflow for deployments
  • Immutable infrastructure patterns
  • Comprehensive testing in staging before production
  • Automated rollback mechanisms
  • Security scanning in CI pipeline

Future Roadmap

  • Multi-cluster support for disaster recovery
  • Advanced cost allocation and chargeback
  • Machine learning-based autoscaling
  • Enhanced developer experience with CLI tools