The practical implementation of many quantum algorithms known today is believed to be limited by the coherence time of the executing quantum hardware and quantum sampling noise. Here we present a machine learning algorithm, NISQRC, for qubit-based quantum systems that enables processing of temporal data over durations unconstrained by the finite coherence times of constituent qubits. NISQRC strikes a balance between input encoding steps and mid-circuit measurements with reset to endow the quantum system with an appropriate-length persistent temporal memory to capture the time-domain correlations in the streaming data. This enables NISQRC to overcome not only limitations imposed by finite coherence, but also information scrambling or thermalization in monitored circuits. The latter is believed to prevent known parametric circuit learning algorithms even in systems with perfect coherence from operating beyond a finite time period on streaming data. By extending the Volterra Series analysis of dynamical systems theory to quantum systems, we identify measurement and reset conditions necessary to endow a monitored quantum circuit with a finite memory time. To validate our approach, we consider the well-known channel equalization task to recover a test signal of N_ts symbols that is subject to a noisy and distorting channel. Through experiments on a 7-qubit quantum processor and numerical simulations we demonstrate that N_ts can be arbitrarily long not limited by the coherence time.