Seesa Code · an AI coding agent on your own hardwareSoftware work, on your own hardware.
Seesa Code works in your repository. It reads the code, makes the change in a sandbox and shows you the diff. Nothing is committed or pushed until you approve.
- For
- Engineers and the people who work with them
- Runs on
- The same installation as your Seesa
- Status
- Early access
- Price
- Free
One session per task.
Point Seesa at a project and describe the change. Each task gets its own workspace and its own conversation, which is still there when you come back.
Read the change before it exists anywhere else.
A file tree, a live diff of what changed and the history of the branch, beside the conversation that produced them.
Commit, push and open the pull request, on approval.
Seesa prepares the commit and the pull request. An approval card shows exactly what will be pushed, and nothing leaves until you say yes.
Your code stays on your hardware.
By default Seesa Code uses the models already running in your installation. An administrator can allow cloud models with your own keys, and every session then says clearly that code leaves the box.
Shown with Tarnhollow, a demo company. Names and figures are illustrative.
Questions, answered.
What is Seesa Code?
Seesa Code is an AI coding agent that runs on your own hardware. It works in your repository inside a sandbox, shows the diff, and commits or pushes only with your approval.
Does our source code leave the building?
Not by default. Seesa Code uses the models in your installation. Cloud models are off unless an administrator turns them on.
Who is it for?
Engineers, and colleagues who need a change made without learning the tooling. The same Seesa that answers questions about the company can take on the code task.
The Seesa family.
See Seesa Code on your own repository.
Tell us about your team and the work you would like Seesa to take on.