Prototype to production
in one file.
Got an idea? Ship it. Launchpad gives every prototype a real URL, real infrastructure,
and a path to production — all from a single .launchpad.yaml.
▋
Got an idea? Ship it. Launchpad gives every prototype a real URL, real infrastructure,
and a path to production — all from a single .launchpad.yaml.
Launchpad is designed for teams that move fast. Hack on a prototype over the weekend, demo it on Monday with a real URL, and scale it to production when it sticks — without re-platforming. No Dockerfiles. No Helm charts. No infrastructure tickets. Just push your code.
Go from idea to running app in minutes, not days.
Push a .launchpad.yaml and get a running prototype. Runtime auto-detection, Dockerfile generation, and packaging happen automatically.
Every pull request gets its own URL. Share prototypes with stakeholders before merging. Environments auto-cleanup when PRs close.
Start with defaults, scale when ready. Same config file works for a hackathon demo and a production service. No re-platforming required.
Even prototypes get isolated namespaces, config validation, and secrets management. No shortcuts that bite you later.
Stream logs from the CLI on day one. Datadog metrics and traces are wired up automatically. Debug prototypes like production apps.
A tiny GitHub Actions stub is all you need. Reusable workflows handle build, push, and deploy. Spend time on your idea, not your pipeline.
Any team that needs to ship fast and iterate.
Deploy your hack in minutes. Get a shareable URL before the demo.
Validate ideas with real infrastructure. No "works on my machine" demos.
Ship dashboards, bots, and automation scripts without ops overhead.
Deploy LLM-powered prototypes with a Python or Node backend in seconds.
From idea to live prototype in 3 steps.
Run launchpad init in your project. It detects your runtime and generates everything you need.
name: my-prototype
runtime: python
team: dai
config:
env:
LOG_LEVEL: infoArgoCD detects the repo, clones your code, installs dependencies, and runs it. No CI needed for source-based runtimes (Node, Python, Go). For runtime: dockerfile, push your image to GHCR and the cluster pulls it via ECR.
# runtime: node/python/go — just push, no CI needed
git push origin main
# runtime: dockerfile — push image to GHCR
docker push ghcr.io/DND-IT/my-app:latestYour app is live with a real URL. Share it, get feedback, iterate. When it's ready for production, it already is.
$ launchpad status my-prototype
Name: my-prototype
Runtime: python
Health: Healthy
Replicas: 1
URL: my-prototype.launchpad.tamedia.techAuto-detected from your project files. Bring any stack.
Ship prototypes from the terminal.
launchpad init
Scaffold .launchpad.yaml and CI workflow. Auto-detects runtime.
launchpad deploy
Trigger a deployment. Shows progress and your prototype URL.
launchpad status [app]
Show health, sync state, replicas, and URL.
launchpad logs [app] --repo org/repo -f
Stream live logs from the latest pod. Use --repo to target a repo from anywhere.
launchpad list --team dai
List all your team's apps and prototypes.
launchpad secret list --repo org/repo
Manage secrets. Use --repo to target any repo without being in the directory.
launchpad completion zsh
Generate shell completion scripts. Add source <(launchpad completion zsh) to your profile.
launchpad login
Authenticate via OIDC device flow.
# Homebrew (recommended)
brew install DND-IT/tap/launchpad
# Go install
go install github.com/DND-IT/launchpad/cmd/launchpad@latestProduction infrastructure, zero complexity.