The increasing penetration of distributed energy resources in modern power systems is intensifying operational challenges related to voltage regulation, particularly in low-voltage networks with heterogeneous assets, limited observability, and strong local coupling between controllable resources. Although centralised optimisation approaches can coordinate corrective actions across multiple devices, they typically require continuous communication with a supervisory controller, full disclosure of local operational data, and increasing computational and communication effort as the number of controllable assets grows. This paper presents DERway as a distributed edge-computing platform designed to support cooperative voltage control through the coordinated execution of a distributed optimisation algorithm across multiple smart gateways. In the proposed approach, each DERway gateway is deployed close to a subset of local DER assets and executes an ADMM-based control routine using local measurements, local constraints, and limited exchanged variables. Rather than relying on a fully centralised decision layer, the gateways cooperate iteratively to converge towards a coordinated solution that mitigates voltage violations while preserving the decentralised nature of control. DERway provides the practical execution environment for this distributed coordination through modular local services, device-level communication interfaces, internal data handling, and edge-side processing capabilities. The platform supports acquisition of local measurements, execution of the optimisation routine, and exchange of coordination variables between gateways, enabling distributed control to be deployed close to the physical assets. The proposed framework was evaluated in a smart-grid control context involving multiple cooperating gateways and voltage-regulation scenarios with distributed controllable resources. The results show that the distributed edge-based approach can successfully mitigate voltage violations through coordinated local decision-making, while reducing dependence on centralised control and avoiding the need to disclose full local datasets to a single supervisory entity. The work therefore demonstrates that smart DER gateways can evolve beyond interoperability functions and act as edge-native optimisation nodes for scalable and privacy-aware operation of distributed energy systems.