JF
As you rightly say, on Concorde, at speeds above M1.3, variable position ramps, situated inside the engine intakes become active and are used to regulate the quantity and speed of the intake air arriving at the compressor. They do this by focusing a series of shock waves on the lower lip of the engine intake, through which the (supersonic) intake air is forced to travel, being decelerated and compressed during this process, before arriving at the compressor face at about M0.5.
Likewise, at the exhaust end, the exhaust gas has to be accelerated to very high speed to produce the required thrust, which is done by a system of primary and secondary nozzles forming an efficient convergent/divergent nozzle system.
One significant and rather unusual difference to bear in mind when considering supersonic airflow is that a convergent nozzle will slow and compress a supersonic airflow whereas a divergent nozzle will slow and compress a subsonic airflow.
The design of the intake and exhaust systems of a supersonic engine is critical, and plays a major part in the efficiency of the engine. Due to the efficiency of the design of the intake and exhaust systems on Concorde, at M2.0, roughly 25% of total thrust is produced by the intake system, with another 25% being produced by the exhaust system.
This leaves the core engine to produce only 50% of the total thrust required in cruise, well within its capabilities without requiring the use of reheat (with its attendant high fuel flow) and enables flight at M2.0 at sustainable fuel flow rates.