Your app, in production

Connect your repo. Tael takes it from there.

Tael puts your app in production and keeps it there — an AI that deploys, watches, and maintains your software.

Free to get live. No credit card.

Lovable writes the code. Cursor writes the code. Claude writes the code. Then what?

Writing software used to be the expensive part. That cost just collapsed.

Running software — keeping it up, patched, fast, and paid for — costs exactly what it did last year.

Every new app ships with that gap. Someone has to carry it.

Day 1 is a feature. Day 2 is the product.

How it works

From repo to production in one sitting

Three steps. The last one never ends — that is the point.

01

Sign in with GitHub

One button. I only ask for access to the repos you choose.

02

Tael reads your repo and shows you the plan

I work out what your app needs and put the plan in front of you before anything runs.

03

You're live, and Tael is watching

A real URL on real infrastructure, with me on call from the first minute.

Day 2

The 2am List

The things that break after launch — and what Tael does about them.

SSL certificate expiry
What happens
The certificate that keeps your site on https quietly runs toward its expiry date.
What Tael sees
I track the expiry date for every domain you run and flag anything inside sixty days.
What Tael does
I renew the certificate automatically and verify the new one is actually being served.
What you see
A line in your feed: renewed and verified, weeks before it could have mattered.
Out-of-memory crash at 2am
What happens
Your app’s memory use creeps up for days, then the process is killed mid-request.
What Tael sees
I catch the crash, pull the logs around it, and find the pattern that caused it.
What Tael does
I restart your app to stop the bleeding, then write up a fix with the evidence attached.
What you see
One message with the diagnosis and an Approve button — it can wait until morning.
Dependency vulnerability
What happens
A package your app depends on discloses a security hole.
What Tael sees
I match the advisory against the code your app actually runs, not just what is installed.
What Tael does
I open a pull request with the patched version, and your checks prove it still builds.
What you see
A PR waiting in the morning — merging it is the approval.
Disk filling up
What happens
Logs and uploads grow until your app runs out of room, usually at the worst moment.
What Tael sees
I watch the trend, not the threshold — I know it will be full in nine days, not that it is full now.
What Tael does
I propose more space with the exact monthly cost printed next to it.
What you see
A one-line decision — "Add 20 GB for €2 a month?" — approved in one tap.
Traffic spike
What happens
You get featured somewhere and traffic triples inside an hour.
What Tael sees
I watch the surge in real time — response times, error rates, and how much headroom is left.
What Tael does
I add capacity inside the budget you set, then wind it back down when the wave passes.
What you see
A report after the fact: what spiked, what it cost, and that nobody saw an error page.
Broken deploy
What happens
A release that passed its checks still misbehaves once real traffic hits it.
What Tael sees
I compare error rates before and after every deploy, so a bad one stands out in minutes.
What Tael does
I hold the rollout automatically and keep the last good version serving traffic.
What you see
"This deploy looks wrong — here is the diff. Roll back?" One click either way.
Forgotten environment variable
What happens
New code reads a setting that was never added to production.
What Tael sees
I compare what your code expects with what production actually has — before the deploy, not after.
What Tael does
I stop the deploy and tell you exactly which value is missing.
What you see
A clear message at deploy time instead of a mystery outage at 3am.
The morning after

You slept. Nothing else did.

Tael keeps a plain-language feed of everything it saw and did. No dashboards to learn — just what happened, when, and what it cost.

Entries in lila are Tael’s work. Entries in jade are yours. That distinction runs through the whole product.

While you were awayEvery action on the record, with evidence.
live
  • 22:47
    TaelRenewed the certificate for app.acme.dev — 62 days early.
  • 01:19
    TaelBlocked lodash 4.17.20 — known vulnerability. PR #41 is open and your checks already pass.
  • 08:30
    YouYou merged PR #41. The patched version is live.
  • Tue 14:00
    TaelTraffic doubled Tuesday 14:00; scaled to 3 instances, back to 1 by 16:20. Cost: €0.11.
  • Wed 03:12
    TaelCleaned 1.4 GB of old build artifacts. Disk trend is flat again.
  • now
    Tael14 checks a minute, all quiet. Nothing needs you.
Trust

Fast because it’s constrained

Tael moves quickly precisely because the boundaries are explicit.

Everything is a change you can see

Every action I take is a commit or a pull request in your repo, with the evidence that justified it. Nothing happens off the books.

You choose the leash

Ask-first by default. Pre-approve the categories you trust me with, each with a budget. And there is a global pause button that always works.

Some things are always yours

Deleting data, spending over budget, anything touching your domain — always a human decision. I do not have a mode that overrides you.

Your repo contains everything needed to run this app anywhere. Leave whenever you want — it all still works.

Pricing

Hosting is cheap. We price it that way.

You’re paying for the engineer, not the servers.

Your app deserves a production.

Connect your repo and I’ll show you the plan before anything runs.

Continue with GitHub