Your strategies run on our compute, not your laptop.
Backtests, paper runs, and live runs execute on managed cloud infrastructure. Each run is isolated, monitored, and cleaned up for you. You write the strategy; Periscøpe handles the servers, the runtime, and everything around the run.
Run strategies without running infrastructure.
No infrastructure to run
No servers to provision, no schedulers to babysit, no runtime to keep alive. Launch a run and it executes on managed cloud compute.
Isolated by default
Each strategy runs in its own locked-down sandbox, with no network, a read-only filesystem, and strict resource limits, so one run can't reach or affect another.
Long-running paper and live
Paper and live runs are real-time jobs that keep running until they finish, you kill them, or they error, and they are monitored the whole time.
Cleaned up after
Runs are supervised and torn down when they're done, so nothing is left hanging and you're not paying attention to plumbing.
Launch a run; the platform does the rest.
Choose a strategy version, set parameters and a mode, and launch. Periscøpe places the run on managed compute, wires up the data feed and broker for the mode, and supervises it for you.
- Backtest, paper, or live from one place
- Managed compute provisioned per run
- Data feed and broker wired up by mode
- Nothing to deploy or keep alive
Watched from launch to teardown.
While a run executes, its state, positions, orders, fills, and logs are tracked and surfaced. When it ends, whether completed, killed, or failed, the compute is cleaned up and the record stays.
- Live run state and metrics
- Positions, orders, fills, and ledger
- Strategy logs and classified issues
- Automatic cleanup when a run ends
What managed compute is for.
Running strategies you don't want to babysit.
If you'd rather spend your time on strategies than on servers, schedulers, and keeping a runtime alive, managed compute takes the operational load off you.
Wanting to own the host.
A managed platform runs your strategies on its infrastructure by design. If you specifically need to control your own servers and deployment, that's a different trade-off.
Managed compute: questions.
Where do my strategies run? +
On managed cloud compute (AWS). Backtests run as short-lived jobs; paper and live runs are long-running real-time jobs. You don't provision or manage any of it.
Is my strategy code isolated? +
Yes. Strategy code runs in a hardened sandbox, with no outbound network, a read-only filesystem, a non-root user, and memory and process limits, and reaches the simulation only over a controlled channel.
How are paper and live runs hosted? +
They run on the same managed compute as backtests, as long-running real-time jobs that execute until they complete, you stop them, or they hit an error. The platform monitors them throughout.
Do I need to manage servers, schedulers, or deployments? +
No. That's the point of a managed platform: the runtime, scaling, and run lifecycle are handled for you. You write and publish the strategy; Periscøpe runs it.
What happens when a run finishes? +
The run is supervised to completion and its compute is cleaned up. The run record (config, orders, fills, positions, ledger, issues, and logs) stays available for you to review.
Run strategies without running servers.
We onboard in focused cohorts. Tell us what you trade and how you work. Live access is rolled out to select users on request.