Self-hosted job orchestration

Job orchestration you actually own

Schedule scripts and commands, run them on machines you control, and know the moment something breaks — with dependency chains, auto-retries, and failure alerts.

free to start · no credit card · 2 agents included

12jobs
3/3agents online
148runs · 24h
0failed · 24h

What it does

Everything a pile of cron jobs can't

If you can run it from a command line, ThinkTorch can schedule it, chain it, retry it, and tell you when it breaks.

cron

Real scheduling

By the minute, hour, day, or a full cron expression — with the next run time shown for every job.

agents

Your own machines

Jobs run on hardware you control via a small outbound-only agent. Commands and data never leave your infrastructure.

chains

Chains & retries

Trigger jobs on success or failure, auto-retry the flaky ones, and stop runaway runs with a timeout.

alerts

Know it broke

A live dashboard of failures and activity, plus an email the moment a job breaks.

Where it fits

Past a pile of cron jobs, nowhere near Control-M

Cron / Task Scheduler Enterprise (Control-M, Rundeck) ThinkTorch
Setup Per-machine, by hand Weeks, often a consultant Install an agent, done
Failure alerts None Yes Email the moment a job fails
Retries & timeouts DIY scripting Yes Built in, per job
Job dependencies Cron can't Yes Chain on success or failure
History & logs Scattered or none Yes Every run, in one place
Runs on your machines Yes Usually their infra Yes — agent is outbound-only
Price Free $$$$ + procurement Free to start, from $19/mo

Getting started

Up and running in three steps

01 — connect

Connect an agent

Run a tiny program on any machine you own. It reaches out to us — no inbound ports to open.

02 — define

Create a job

Define the command to run, a schedule if you want one, and how many times to retry on failure.

03 — watch

Watch it run

Follow live status and logs, see failures at a glance, and get emailed if something breaks.

Use cases

What can you do with it?

If you can run it from a command line, ThinkTorch can schedule it, chain it, retry it, and tell you when it breaks. A few things teams use it for.

backups

Backups you can trust

Back up databases or files on a nightly cron — and get emailed the moment one fails, instead of finding out when you need the backup.

pipelines

Pipelines that wait their turn

Chain extract → transform → load. Each step runs only when the previous one succeeds; if one fails, the chain stops and you're told.

health

Uptime & health checks

Ping a URL or service every few minutes. If it doesn't come back healthy, the run fails and an alert lands in your inbox.

reports

Scheduled reports

Generate and send reports on a schedule, and auto-retry the flaky steps so a momentary blip doesn't skip a day.

cleanup

Housekeeping at scale

Rotate logs and clear temp files across every server you run an agent on — one schedule, all your machines.

webhooks

Glue between systems

Trigger a deploy, refresh a cache, or ping a chat channel via webhook the moment an upstream job finishes.

Where it runs

Runs on your machines

The agent runs wherever your jobs need to run. Your commands and their output never leave your hardware.

Windows — available now Linux — on the roadmap macOS — on the roadmap

Pricing

Simple, honest pricing

Start free, upgrade when you outgrow it. Every plan has every feature — you only pay for scale.

Free
$0
Kick the tires - every feature, capped resources.
  • 2 agents
  • 1 environments
  • 5 jobs
Start free
Starter
$19 /mo
or $190/yr — 2 months free
For a solo operator or a tiny setup.
  • 5 agents
  • 2 environments
  • 20 jobs
Get Starter
Business
$149 /mo
or $1490/yr — 2 months free
For orgs that need room and (soon) governance.
  • 40 agents
  • Unlimited environments
  • Unlimited jobs
Get Business

Prices in USD. Every plan includes every feature — cancel anytime.

FAQ

Questions, answered

Do I have to open any firewall ports?

No. The agent only makes outbound connections to ThinkTorch and asks for work, so there's nothing inbound to expose.

Where do my jobs actually run?

On machines you own, via the agent you install. Your commands and their output never leave your infrastructure — we store the schedule, status, and any logs you choose to send, and nothing more.

What can a job be?

Anything you can launch from a command line: a PowerShell or Python script, a batch file, or any executable, with the arguments you give it.

What happens when a job fails?

It can auto-retry a number of times you set, it emails your notifications address, and it shows on your dashboard's last-24-hours view with a link straight to the log.

Is there really a free plan?

Yes — free forever, no credit card. You only move to a paid plan when you need more agents, environments, or jobs. Every plan has every feature.

Can my whole team use it?

Yes. Invite teammates and give each a role per environment (Viewer, Operator, Editor, Admin). It's one workspace and one bill — there's no per-seat charge.

Which operating systems are supported?

The agent runs on Windows today. It executes whatever commands you'd normally run there — cmd, PowerShell, Python, and so on.

Can I cancel anytime?

Yes, from the Billing page. Downgrading or canceling never deletes anything you've already created.

Create a workspace in under a minute

Free forever, no card. Every plan has every feature — you only pay for scale.

Create your workspace