Abstract : Modern web applications face increasing demands for responsive performance, scalable architecture, and robust privacy safeguards. This paper investigates methods for optimizing client-side data processing and storage by leveraging modern technologies—IndexedDB, Web Workers, TensorFlow.js, and WebAssembly—in tandem with AI techniques. We explore how these tools address performance constraints arising from the browser’s limited computational resources while also mitigating privacy concerns by reducing data transfers to remote servers. We present empirical results on processing one million data records in different scenarios (with and without AI), highlighting trade-offs in execution time, memory usage, and clustering accuracy. Furthermore, we analyze the economic implications of shifting computational workloads from central servers to client devices, demonstrating potential cost savings in high-frequency data processing environments. The study concludes by offering concrete recommendations for developers to create efficient, secure, and intelligent web applications that can handle large-scale data on the client side.

Keyword : Client-side Data Processing, Cost-efficient Web Applications, TensorFlow.js, WebAssembly.