The current service recipients of the Xinjiang Meteorological Big Data Cloud Platform are mostly based on numerical forecasting models and multisource fusion analysis of meteorological intranet business systems. The integration with the mobile internet is not high, and most of the frontline observation data are directly extracted and processed without being visually displayed through media. The basic stations also lack the means to access the operational status of automatic stations and local refined forecasting and warning on mobile devices. Therefore, to enhance the quality of meteorological services and meet the mobile visualisation needs for meteorological disaster prevention and reduction, as well as to improve the monitoring and warning capabilities of small and mediumscale weather disasters at the basic level in Xinjiang, there is urgency to develop a professional meteorological service APP. The Tian Shan Weather APP breaks through the heterogeneous barriers between the Xinjiang meteorological intranet and the mobile internet at the network level, fully utilises the existing digital achievements of meteorological business in the “cloud + end” business model. It processes and displays meteorological data such as ground observations from automatic weather stations, gridbased actual conditions, radar echoes, and disaster warnings. The key technologies are as follows: using multisource data processing and storage technology to process and reprocess raw data and standardised storage products; using service interfaces that inherit the technical characteristics of the largescale data cloud platform and achieve millisecondlevel data calls; using the SSM development framework based on inversion of control to separate APP transaction management from main business logic; using a data interaction mode based on the Restful architecture style for crossplatform development. At present, the Tian Shan Weather APP can provide realtime weather conditions, refined forecasts, warning details, and radar mosaics based on the user’s location in Xinjiang according to the user’s needs. It can also customise the calculation and analysis products of ground actual condition elements and statistical value elements for meteorological research purposes. Through simulated testing of the data cloud platform service interface and the original CIMISS system interface, it is found that under 100 concurrent accesses, the data cloud platform interface is 5.7 times faster than the original CIMISS interface, and under 200 concurrent accesses, it is 5.3 times faster. This shows that the APP inherits the high carrying capacity of the data cloud platform service interface. The APP is being widely promoted and used by meteorological departments at the provincial, municipal, and county levels, as well as in related industries such as civil aviation and the air force in Xinjiang. It has been widely recognized and has achieved significant results in the emergency support of weather automatic stations and meteorological disaster prevention and reduction services throughout Xinjiang.