Generative AI and Cloud Data Engineering for Business Intelligence
Аннотация
The GenAI and cloud-native data engineering confluence is revolutionizing the BI landscape to allow for real-time analytics, at scale, but closer to the human scale. This paper reviews the technical underpinnings, experimental validations, and system architecture of accommodating LLMs in cloud data platforms like Snowflake, BigQuery, and Azure Synapse. We evaluate the effectiveness of GenAI for augmenting data-to-insight pipelines from natural language querying, to prediction, forecasting and report generation, across ten key studies developed over 2019–2023. We also propose a unified theoretical framework, provide benchmark studies and experimental findings, which reveal GenAI's shift to a general-purpose, domain-agnostic, transformative utility but the limitation of practical cost, interpretability, and ethical deployment. Finally, we propose future research directions focusing on explainability, domain adaptation, and sustainable architectures for enterprise level adoption.