About
We build software that keeps people in charge of their AI.
Runlance started with one rule: an agent can do the work, but a person decides what leaves the workspace.
Mission
Make AI work visible.
Agents can take on real work. Teams only trust them when they can see each step and decide what happens with the result.
That is why every Runlance run ends with a person. We design for review first, and for speed second.
AI-generated image.
Story
How Runlance came together.
From a weekend prototype to the workspace where agents and reviewers work side by side.
2024
A weekly report
The idea starts with one team tired of writing the same status update every Friday.
2025
Review first
The first version ships with one rule: nothing leaves without a person approving it.
2026
One workspace
Agents, models and workflows move into one shared workspace for the whole team.
Values
What we hold on to.
Visible work
If an agent did it, you can see how.
People decide
Software proposes. A person approves.
Calm software
Fewer notifications, clearer results.
Team
The people behind the product.
A small team of product, engineering and design people. Everyone ships, everyone reviews.
MK
Maya Kern
Co-founder, Product
JL
Jonas Lind
Co-founder, Engineering
AS
Aylin Sar
Design
TB
Tom Berg
Customer success