The field is moving towards more efficient and scalable solutions for edge computing and time series analysis. Researchers are exploring novel architectures and models that can handle the complexities of real-time data processing and reduce the latency associated with traditional cloud-based approaches. Notable developments include the use of serverless architectures, adaptive normalization techniques, and transformer-based models to improve the accuracy and efficiency of time series forecasting and analysis. Some papers have presented innovative solutions to specific challenges, such as mitigating the cold start problem in FaaS architectures and generating domain-aware captions for time-series images. Noteworthy papers include: Trabant, which introduces a serverless architecture for multi-tenant orbital edge computing, Fast-Powerformer, which proposes an efficient and lightweight model for mid-term wind power forecasting, and TADACap, which presents a retrieval-based framework for generating domain-aware captions for time-series images.