The real-time requirements of closed-loop neural interfaces demand an innovative approach to co-design of hardware and software solutions. Signal processing, decoding and control algorithms must be applied to high bandwidth, multichannel data using wearable or implantable hardware operating within strict power and size constraints. We will engage with cutting-edge research into embedded systems, wireless and cloud architectures, as well as emerging approaches such as neuromorphic computing to leverage engineering advances from different domains in CNITs. This WS will be informed by the computational and experimental work undertaken within the broader network as well as user involvement to define specifications and requirements for closed-loop devices.