Routing core
Campaigns, numbers, buyers, and targets — the decision engine that directs every incoming call to the right endpoint.
- Rules and targets per campaign
- Buyer endpoints and priorities
- A routing history for every call
The open-source core of Callfinix
Self-host the core.Your server. Your call data.
Open sourceLicense TBDSelf-hosted
The routing core, scheduler, and dashboard behind Callfinix — running on your own infrastructure, with every call staying on your side of the wall.
The repository is being finalized — the GitHub URL, license, and setup commands on this page are placeholders until it lands.
Routing coreSchedulerDashboard
What’s in the repo
Three pieces.
One core.
Everything a call operation needs to route, keep time, and show its work — without a vendor in the middle.
Scheduler
The always-on clock. It drives the recurring work so a self-hosted stack behaves like an operated one.
- Campaign schedules and windows
- Webhook delivery with retries
- Nightly upkeep and reports
Dashboard
The picture of the operation you know from Callfinix, served from your own server instead of ours.
- Calls, routes, and outcomes together
- Partner and payout views
- Runs where your data runs
Dashboard screenshot · Placeholder
See the dashboard here when the core ships.
A real screenshot will replace this space once the repository is finalized and imagery is approved.
Zero to running
Seven steps.
Then place a call.
Command-first, like the repo README. Clone, configure, start — and dial through your own stack.
The repository is being finalized — the GitHub URL, license, and setup commands on this page are placeholders until it lands.
Clone the core
One checkout brings the routing core, the scheduler, and the dashboard.
git clone <repository-url> && cd callfinix-corePoint it at your environment
Copy the example environment, then set your database and telephony provider keys.
cp .env.example .envStart the services
Core, scheduler, and dashboard come up together on your host.
docker compose up -dPrepare the database
Apply the schema so campaigns, numbers, buyers, and routes have a home.
docker compose exec core callfinix migrateAdd your first number and buyer
Register a campaign number and a buyer target — the two pieces a route needs.
callfinix numbers:add --campaign firstPlace a test call
Dial the campaign number and watch the routing core pick the buyer.
callfinix routes:test --campaign firstWatch it land
Open the dashboard: the call, its route, and its result — on your screen, on your server.
open http://localhost:8080
Always on
Keep it running.
Day after day.
A self-hosted core still has to behave like an operated one. The scheduler carries the recurring work; these habits carry the rest.
Know it’s up.
The scheduler publishes a health heartbeat. Point your monitor at it and page a human when it goes quiet.
Survive the restart.
Run the stack with a restart policy and the core, scheduler, and dashboard come back after a reboot — no hand-holding.
Upgrade in place.
Pull the latest core, rebuild, migrate. Your campaigns, routes, and history move forward with it.
When it misbehaves
Symptom.
Cause. Fix.
The five stops you’re most likely to hit on the way up — and the move that clears each one.
| Symptom | Cause | Fix |
|---|---|---|
| Calls land, but no route fires. | The campaign has no active buyer target. | Open the routing timeline in the dashboard, add or re-enable a target, and dial again. |
| Scheduler stopped after a reboot. | The services run without a restart policy. | Set restart: unless-stopped in the compose file and bring the stack back up. |
| Dashboard loads, but shows no data. | The core can’t reach the database. | Check the database URL in .env, then restart the services. |
| Webhooks aren’t delivering. | Outbound HTTPS is blocked from the host. | Allow egress for the core and watch the pending deliveries retry. |
| A test call never reaches the buyer. | The number isn’t attached to the campaign. | Re-check the number’s campaign assignment in the dashboard and place the call again. |
Questions
Before you clone.
What does the open-source core cost?
Nothing. The self-hosted core is free to run, study, and modify. The paid Callfinix service is a separate, optional layer on top of it.
How is this different from Callfinix?
The core gives you routing, scheduling, and a dashboard on your own server. Callfinix adds AI agents that run the operation — setup, triage, routing, and reporting — plus managed infrastructure and support.
Does my call data leave my server?
No. Calls reach your core through your telephony provider, and the recordings, transcripts, and reports stay where you run them.
Can I run the core and still use Callfinix?
Yes. Route some campaigns through your self-hosted core and others through Callfinix, or move campaigns between them as the work changes.
What’s the license?
The final license lands with the public repository — the badge at the top of this page carries it as soon as it’s chosen.
Open core, paid service