Sandboxes
First sandbox run
The first SAVRN Cloud sandbox walkthrough uses built-in text inspection or JSON validation tasks with fictional inputs. Follow the sample lifecycle, result and stop controls to understand the intended workflow without treating the browser operation as an isolated compute environment.
SAVRN Cloud · Coming soon · Public preview · Browser-local simulation · No live compute or payments
Complete the local walkthrough
Open Sandboxes and choose New sandbox. Select a built-in task: text/CSV shape, JSON validation or input-shape summary. Choose a 30- or 60-second limit and 256 or 512 MB. Use a short fictional input. These are local demonstration settings, not allocated host resources.
Choose Run task, inspect output and activity, then use Stop task where appropriate. Export run preserves the sample result. Confirm the recorded transition before creating another resource. Arbitrary code, GPU execution and network access are not supported by this local workflow.
What success means
The walkthrough succeeds when you can understand the requested resource, observe the local lifecycle, inspect its simulated result and identify how to stop it. A success label demonstrates the interface flow only. It does not prove a real command executed, a network rule blocked traffic or a supplier charge was reconciled.
Before connected services launch
A real first-run guide must use a qualified image, explicit resource/time ceilings and a permitted data category. It must explain input transfer, secret scope, output retention and cancellation. Verify cleanup through the actual scheduler/provider and preserve a resource-time receipt. Test timeout, worker failure and retry behavior. Publish these instructions only after the same bounded task completes against the released execution service.
Build toward the work that matters.
Tell SAVRN what your institution needs to run, who reviews the results and where its data must stay. That workload defines the next service to qualify.
Discuss an AI project