10171210AI Driven No Code Network OptimisationClosedFast Start ResponseInnovate UKDiversity of applications using wireless (both mobile and WiFi) networks is increasing alongside the demands users are placing on network capacity and performance. Within both private (e.g. smart factory) or public (e.g. urban space) this pressure leads to increased specialised configuration of networks. This customisation are often dependant on the vendor specific hardware and software interfaces that they are built on. They require some specialist knowledge to use them and are often out of bounds for the non technical user. Regulation in the networking technologies domain is dominated by universal service obligations. Third party online tools are increasingly common to assist users when checking service issues. These issues lead to support requests which in the broadband sector are largely handled by human call centres. BT moved all customer service calls back to the UK and Ireland, citing improved first-contact resolution and customer satisfaction. Within the mobile network domain this support is increasingly being replaced by automated chat bots. Giffgaff and Smarty have no phone-based support at all. Within private networks support is often supplier / vendor specific related to service contracts. This can increase costs to the end user and tie them to vendor specific hardware. For private connectivity it is feasible to use off the shelf hardware and utilise open standards such as Open RAN and Open Wireless Router (OpenWRT). Here support can be found within community forums. However, in such cases it is increasingly possible to utilise Large Language Models for network support. This innovation within Artificial Intelligence (AI) has the potential also in public network. This project will fine tune large language models and test them against common support issues in public and private networks. Utilising open technology the LLMs will link to automated configuration of networking hardware to help solve issues using no code approaches. This LLM approach to network support supports the automation trend in emerging mobile network support. The project will target expected growth within this support technology and target public and private network providers. The project will deliver a prototype that can be used to exploit the business opportunity targeting both domestic and commercial wireless network providers. Thus, project will develop the AI LLM support solution as a prototype testing it in both public and private scenarios.