Aquaculture is expanding globally, valued at £159 billion in 2022, with Atlantic salmon dominating finfish production at 3.5 million tonnes annually. As the industry grows, more sophisticated technology is needed to monitor farms and ensure their sustainability. Using behaviour as a non-invasive form of monitoring, in combination with artificial intelligence and machine learning, can allow for higher control over farm management. For instance, the development of algorithms to analyse fish behaviour related to feeding may be used to fully automate the feeding process and reduce environmental and economical waste. The goal of this study is to identify changes to Atlantic salmon (Salmo salar) behaviour related to responses to environmental conditions (e.g., currents, oxygen, temperature), operational procedures, therapeutic mechanical or medicinal treatments or health status of the fish (AGD, PGD, sea lice burden). Therefore, it is essential to understand how the fish are distributed within a cage, and where best to place cameras to gather reliable behavioural data. For this study, 5 cameras were deployed at a Scottish Atlantic salmon farm consisting of 10 cages, each 100 m in circumference and ~15 meters in depth. The cameras were deployed in one cage in the following orientation: 3 down the centre (4 m, 9 m, 14 m), 2 at 9 m on the inner and outer areas of the cage, respectively. An algorithm was created by Observe Technologies to process video footage from these cameras and transform it into behavioural data useful for farmers (e.g., activity, speed, schooling). First analysis shows that with respect to activity, the camera in the centre of the cage at 9 m is the most accurate for tracking feed times, indicating that this is a valuable placement of the camera to gather data on feed behaviour. Additionally, there is significantly higher activity on the side of the cage closer to the inner farm compared to the opposite, more exposed side, and more activity at the bottom of the cage compared to the surface. This suggests that the fish are congregating in these two locations, possibly to avoid stronger currents and for fear of predation at the surface, respectively. Further work is underway to investigate how environmental or health conditions affect this positioning to ultimately automate feeding and reduce feed waste.