This paper presents an energy-and-area-efficient ACcoupled front-end for multichannel recording of wideband neural signals.The proposed unit conditions local field and action potentials using an inverter-based capacitively-coupled low-noise amplifier, followed by a perchannel 10-bit asynchronous SAR ADC. The adaptation of unit-length capacitors minimizes the ADC area and relaxes the amplifier gain so that small coupling capacitors can be integrated. The prototype in 65nm CMOS achieves 4× smaller area and 3× higher energy-area efficiency compared to the state of the art with 164 µm×40 µm footprint and 0.78 mm 2 ×fJ/conv-step energy-area figure of merit. The measured 0.65 µW power consumption and 3.1 µVrms input-referred noise within 1Hz-10kHz bandwidth correspond to a noise efficiency factor of 0.97. Index Terms-AC coupling, front-end, inverter-based amplifier, monotonic switching, neural recording, successive approximation register analog-to-digital-converter, unit length capacitor.