Built for real volume
Quillo routes forty thousand tickets a month through Hatch queues and SLAs, including the escalation rules it used to keep in a spreadsheet.
Hatch gives support and engineering teams an open-source help desk they can shape, extend and self-host, with AI that answers from your own docs.

Trusted by
+6K others

The problem.
Every agent, add-on and AI resolution is another line on the invoice, and your data lives on their servers.
Shared inboxes and scripts work at ten tickets a day. At a thousand they fall apart.
Stop trading off.
Queues, SLAs, macros and AI from one open toolkit.
Quillo routes forty thousand tickets a month through Hatch queues and SLAs, including the escalation rules it used to keep in a spreadsheet.
Brightmoor pointed Hatch AI at its own docs and changelog. Drafted replies cite the page they came from, so agents check instead of rewrite.
Tallis & Rowe self-hosts Hatch beside its own database. Customer data never leaves its servers, and the per-seat invoice is gone.
Queues, SLAs, custom fields, macros and views come ready to use. Compose the desk your team needs from parts that already hold up at volume.
Extend Hatch with the AI coding tools you already use. Queues, policies and automations are plain TypeScript, so a change is a pull request, not a ticket to a vendor.
Run Hatch on your own servers or in our cloud. Your tickets, customers and AI prompts stay in a database you own, with no proprietary tooling to unlearn.
Skip the clunky UX that custom usually means.
Tickets sit on a board by status. Drag one to move it, the way your team already works.
Everyone sees the same queue update as it happens, so two agents never answer the same customer.
Press ⌘K to reply with a macro, merge duplicates or jump to any queue without reaching for the mouse.
In production.

Fernway runs bike rental support in every city it serves on Hatch, with a queue per city and fleet data inside each ticket.

Calder Health self-hosts Hatch in its own cloud, so patient conversations stay on infrastructure it controls.

Orbit Freight grew its support team tenfold on Hatch without a per-seat bill, and automated most shipment status tickets.
1/3
From the front line
We moved off a per-seat suite in a week. Our queues, macros and SLAs came over as code, and a small change now ships the same afternoon someone asks for it.
Priya Natarajan
Head of Support at Fernway

Any questions?
Yes. The whole desk, including the AI reply engine, ships under AGPL-3.0 on GitHub. You can read every line, run it on your own servers and send changes back upstream.
Most teams forward their support address and have tickets flowing in under an hour. Queues, SLAs and macros come with sensible defaults you can change as you go.
Yes. The importer brings over tickets, customers, tags, macros and help articles, keeps the original IDs as a field and leaves your old tool untouched until you switch the inbox over.
No. Fields, views, queues, automations and macros are all point and click. Developers come in when you want something only code can do.
Every object is defined in TypeScript, so you can add queues, ticket types, actions and integrations in your own repo, review them like any other change and deploy them with the app.
Yes. Hatch exposes an MCP server and a typed API, so assistants can read tickets, draft replies and write extensions, and the AI reply engine can run on the model you choose.
Self-hosting is free. Hatch Cloud is billed per workspace, never per seat: Pro covers unlimited agents and 5,000 tickets a month, and Organization adds SSO, audit logs and a support SLA.