How we think
gittan is built for polyrepo with small repos and simple pipelines, optimised for trunk-based development and flow. These posts explain every deliberate choice, with the reason and the trade-off.
What actually creates software quality
The research is clear. Most of the industry ignores it.
Why your team has no flow
Slow pipelines and PR queues are not friction. They are the reason your team metrics are red.
Features built on research, not feature requests
Accelerate and Team Topologies are our product spec. Not customer wishlists.
Teams own everything. Individuals own nothing.
There is no personal namespace in gittan. Code belongs to teams, because that is how quality happens.
Two permission levels: read everything, own your team
You are either in the team or you are not. There is no in-between.
Why we built gittan
The existing platforms have no direction. We built one that cares whether your team ships well.
It is gated. It is stricter than a PR.
We did not remove the gate. We removed the human-as-bottleneck from the gate.
Small batches and the case for advisory review
When every change is small enough to revert in minutes, pre-approval stops being the only safety net.
Pricing that does not punish collaboration
Per-seat pricing creates perverse incentives. We charge per team.
SSO is a security feature, not a premium add-on
Gating SSO behind enterprise tiers forces small teams into weaker authentication.
Pipelines and policies that help, not block
Automatic pipelines for small teams. Platform guardrails for large ones. Reports over gates.
Security is a pipeline step, not a product tier
Scanning and reporting on every push, for every team. Blocking on critical. Deeper analysis available for teams that want it.
An org changelog that humans can read
What did the team ship this week? Not commits. Not tickets. Features.
Why we integrate with Slack and not Microsoft Teams
Everything about Teams is a project. We chose not to start one.
Why we built for polyrepos
Monorepos solve a tooling problem by creating a platform problem. We solved the tooling problem instead.
Why gittan does not update your dependencies
Dependency updates are engineering decisions. Not automation.
In a gittan pipeline, everything has a pin
No marketplace. Modules pinned by digest. A curated set is paved. Everything else is your explicit responsibility.
Why we ban the latest tag
Every image in a gittan pipeline has a timestamp and a commit SHA. No exceptions.
Why there are no stars, forks, or social features
Stars measure marketing. Team metrics measure engineering.
We build so we do not need a status page
A git host should be invisible. You push, it works. That is the ambition.
Why gittan runs in the EU
Your source code is your business. It should stay in your jurisdiction.
Why gittan is SaaS only
Shared infrastructure, many moving parts, and a deliberate choice.
Why pipeline results belong in the terminal
You are already there. Opening a browser is a context switch.
macOS runners and why they work differently
We can build iOS apps. But the model is fundamentally different from Linux, and here is why.