Kubeletto: The Free, Kubernetes-Powered Way to Deploy Your App in Minutes

If you've been building on the web for more than a few years, you remember the golden age of "just deploy it." git push heroku main and your app was live. No YAML. No clusters. No 2 AM Kubernetes debugging sessions.

That simplicity is exactly why Heroku, and later Railway and Vercel, won over an entire generation of developers - not because they were the most powerful platforms, but because they got out of your way. Then Heroku's free tier disappeared. Costs crept up everywhere. And "just deploy it" quietly turned back into "let's talk about your infrastructure budget."

Kubeletto wants that simplicity back - minus the price tag, at least for now.

What is Kubeletto?

Kubeletto is a serverless container platform built on real Kubernetes-native infrastructure, designed to let you deploy from a GitHub repo or a Docker image and get a live, autoscaled HTTPS endpoint in minutes - with zero cluster management, zero YAML, and zero servers to babysit.

Under the hood, you're getting genuine Kubernetes power: sandboxed build workers, image vulnerability scanning, autoscaling (including scale-to-zero), and traffic-splitting between revisions. On the surface, you're getting a git push and a URL. That gap - real infrastructure, disappearing complexity - is the entire pitch.

The headline: it's free right now, no credit card required

Kubeletto is currently in active beta, and during this beta, usage is completely free. No credit card at signup. You get:

• Up to 20 active service deployments
• 0.1–2.0 vCPU per service
• 128 MiB–2 GiB of memory per service
• A free *.kubeletto.app subdomain with automatic HTTPS
• Secure external database and cache integration
• Full access to the CLI, build pipeline, logs, and console

To be transparent (because good marketing doesn't hide the fine print): this free access is a beta-phase offer. Kubeletto has already published a pay-as-you-go tier that's planned for after beta, priced on actual CPU-seconds, memory-GiB-seconds, and build minutes consumed. There's no fixed date for that transition yet - which means right now is the cheapest this platform will ever be.

If you've ever wished you could go back in time and grab a Heroku free dyno before the world found out, this is that moment.

How deployment actually works

Kubeletto's pipeline is refreshingly linear:

1. Connect a repo or push an image - link your GitHub repository or push straight to a Docker registry.
2. Build container - Kubeletto runs your build inside an isolated, sandboxed worker.
3. Scan & push - the resulting image is vulnerability-scanned and stored in a private registry.
4. Deploy HTTPS URL - you get a live, autoscaled endpoint with SSL already configured.
5. Monitor - stream logs and watch metrics in real time from the console.

Or skip the console entirely and drive it from your terminal:

curl -fsSL https://releases.kubeletto.com/cli/install.sh | bash
kubeletto deploy
kubeletto logs
kubeletto rollback
kubeletto env set

From git push to a production URL, Kubeletto advertises deploy times under two minutes - with zero servers for you to manage and isolated build workers handling the heavy lifting.

The Cockpit: one console for everything

Kubeletto's management console, called Cockpit, is where the "boring but essential" parts of running an app live:

• Live logs and metrics - streamed in real time, not delayed dashboards
• Traffic splitting - drag a slider to route a percentage of live traffic between two revisions (great for canary releases)
• One-click rollback - every deployment revision is tracked, and reverting is instant
• Environment variables - encrypted secrets stored in a secure parameter vault, not plaintext config files
• Resource monitoring - see exactly how much CPU and memory your containers are actually using

What can you deploy?

Kubeletto supports auto-detection for most popular stacks, a Dockerfile-based build, or your own pre-built image:

• Node.js - Express, Next.js, Remix, NestJS
• Go - Gin, Fiber, or plain standard-library APIs
• Python - FastAPI, Django, Flask
• Static sites - plain HTML/CSS, Vite, Astro
• Docker - any OCI-compliant image
• Others - Ruby, PHP, Java, and more

Basically: if it runs in a container, it runs on Kubeletto.

How it stacks up against Heroku

Kubeletto positions itself directly as a modern successor to Heroku's PaaS model, and the comparison is worth spelling out:

The biggest practical difference for most developers: Heroku charges from day one, while Kubeletto currently doesn't - and Kubeletto's autoscaling means idle side projects and staging environments cost nothing in compute while they're not receiving traffic.

Kubeletto also publishes direct comparisons against Railway and Render if you're evaluating alternatives side by side.

Who is this actually for?

• Indie hackers and side-project builders who want a real production URL without committing to a monthly bill before they've validated an idea.
• Small teams who want Kubernetes-grade reliability (autoscaling, rollbacks, isolated builds) without hiring a platform engineer.
• Developers migrating off Heroku or a pricier PaaS who miss the "just deploy it" workflow.
• Students and learners who want to understand container deployment without paying for a home-lab cluster.

Getting started

# 1. Install the CLI
curl -fsSL https://releases.kubeletto.com/cli/install.sh | bash

# 2. Connect your repo or image and deploy
kubeletto deploy

# 3. Watch it go live
kubeletto logs

Or skip the terminal and sign up directly at console.kubeletto.com/signup - connect a GitHub repo, and you'll have a live HTTPS endpoint before your coffee gets cold.

The bottom line

The platforms that win developer mindshare aren't always the most feature-rich - they're the ones that remove friction between "I wrote code" and "the world can use it." That's what made Heroku's free tier legendary, and it's what Railway and Vercel picked up on afterward.

Kubeletto is making the same bet, backed by real Kubernetes infrastructure instead of a thinner abstraction layer - and for as long as the beta lasts, it's free to try with nothing but a GitHub account.

If you've got a side project sitting in a local repo doing nothing, this is about as low-friction as shipping it gets.

👉 Start deploying free at kubeletto.com

Have you tried Kubeletto yet? Drop your deploy time in the comments - I'm curious how it compares to your current stack.