Skip to main content

​
Penn Labs Infrastructure

Welcome to the Penn Labs Infrastructure documentation. This repository contains all the configurations and tools needed to manage Penn Labs’ cloud infrastructure on AWS, including Kubernetes clusters, databases, networking, and deployment automation.

​
What’s Inside

This repository provides a complete infrastructure-as-code solution for Penn Labs, encompassing:
  • Terraform configurations for AWS infrastructure provisioning (EKS, RDS, VPC, Vault, IAM)
  • Kittyhawk - automated Kubernetes YAML generator built on CDK8s
  • Kraken - GitHub Actions workflow automation using CDK
  • Docker images - custom maintained images for infrastructure services
  • Monitoring & Operations - Grafana dashboards and operational tooling

Terraform

Declarative infrastructure management for AWS resources including EKS clusters, RDS databases, and networking

Kittyhawk

TypeScript-based Kubernetes YAML generator that simplifies deployment configuration with custom constructs

Kraken

Automated GitHub Actions workflows for CI/CD pipelines, built using CDK for Actions

Docker

Custom Docker images for infrastructure services including Django base, database backups, and team sync

​
Key Features

​
Infrastructure as Code

All infrastructure is defined declaratively using Terraform, enabling version control, peer review, and reproducible deployments. The infrastructure includes:
  • EKS Kubernetes Cluster - Production-grade container orchestration
  • RDS PostgreSQL - Managed database with automated backups
  • AWS VPC - Isolated network infrastructure
  • HashiCorp Vault - Secrets management and access control
  • Route53 - DNS management for all product domains

​
Deployment Automation

The deployment pipeline is fully automated through a series of integrated tools:
  1. Code changes trigger GitHub Actions workflows (managed by Kraken)
  2. Kittyhawk generates Kubernetes YAML from TypeScript definitions
  3. Docker images are built, tagged, and pushed to Docker Hub
  4. Kubernetes applies the generated manifests to deploy applications
The infrastructure repository uses a remote S3 backend for Terraform state, ensuring state is shared across the team and protected from local machine failures.

​
Security & Access Control

  • IAM roles scoped per service account for least-privilege access
  • Vault integration for secrets management and rotation
  • Team sync automatically grants access based on GitHub team membership
  • Bastion host for secure cluster access by team leads

​
Monitoring & Observability

  • Grafana dashboards for infrastructure and application metrics
  • Database backups to S3 with automated daily snapshots
  • AWS node termination handler for graceful spot instance handling
This infrastructure supports all Penn Labs products, providing a standardized platform for deploying Django backends, React frontends, and supporting services like Redis.

​
Getting Started

Explore the documentation sections to learn more about each component:

​
Repository Structure

infrastructure/
├── terraform/          # AWS infrastructure definitions
│   ├── modules/       # Reusable Terraform modules
│   └── *.tf           # Resource configurations
├── cdk/               # CDK abstractions
│   ├── kittyhawk/    # Kubernetes YAML generator
│   └── kraken/       # GitHub Actions workflows
├── docker/            # Custom Docker images
├── grafana-dashboards/  # Monitoring dashboards
└── docs/             # Documentation and guides

Build docs developers (and LLMs) love

Get started for freeTalk to us