Answer "should we?" before you launch.
Orbital compute isn't right for every workload. The planning tools evaluate feasibility, model thermal and power constraints, and simulate latency - so you make an informed call, and know the cost, before anything leaves the ground.
Thermal, power, latency, and cost - together.
Any one constraint is easy to model in isolation. Orbital compute lives where they intersect: a plan that fits the power budget but not the thermal one, or meets latency but not cost, isn't feasible. The planning tools evaluate them as one problem and hand back a clear verdict.
One call. Power, thermal, latency, and the verdict.
$ curl "https://api.rotastellar.com/v1/satellites/ISS/feasibility?power_w=400&thermal_w=60" \ -H "Authorization: Bearer rs_demo_rotastellar2026" { "satellite": { "name": "ISS (ZARYA)", "altitude_km": 419.4, "period_min": 92.9 }, "feasibility": { "score": 75, "grade": "B", "viable": true }, "power_budget": { "avg_power_available_w": 530, "peak_solar_w": 789 }, "thermal_budget": { "max_temp_c": 85, "safe_operating_range_c": [0, 50] }, "communication": { "passes_per_day": 4, "round_trip_latency_ms": 93 }, "issues": [ "Workload power demand exceeds 70% of available power", "High thermal dissipation requirement" ] }
A real call to the live API, not a mockup. Feasibility, power, thermal, and latency come back in one response, modeled from the satellite's actual orbital elements. The same envelopes CAE schedules against. Try it with any catalog object or alias (ISS, a NORAD id, and so on).
Model the orbit before you commit.
Should we?
Determine whether orbital compute makes sense for a workload - a clear feasible / infeasible verdict with the binding constraint named.
Reject heat in vacuum
Model radiator capacity, duty cycle, and solar heating cycles so plans stay inside the thermal budget.
Meet the SLA
End-to-end latency modeling with ground-station coverage across LEO, GEO, and inter-satellite hops.
Size the energy budget
Solar-panel sizing, eclipse periods, and battery requirements - the envelope every plan is scheduled against.
Planning feeds execution.
The planning tools share their models with the rest of the platform: the same power, thermal, and link envelopes that prove a workload feasible are the ones CAE schedules against and the Operator Agent executes inside. Plan once; run on-orbit.
Put your first workload in orbit.
Tell us what you want to run and the satellite you want to run it on. We'll come back with a feasible plan - or an honest reason it can't be done yet.