Vol. 10 No. 4 (2026): International Research Journal of Innovative Engineering (IRJIE)

The rapid evolution of digital technologies has transformed the way enterprises collect, process, analyze, and utilize information for strategic decision-making. Artificial Intelligence (AI), cloud-native technologies, and secure digital integration have emerged as foundational pillars for developing next-generation enterprise intelligence systems capable of delivering real-time insights, operational efficiency, and enhanced organizational resilience. AI enables predictive analytics, intelligent automation, and cognitive decision support, while cloud-native architectures provide scalability, flexibility, and cost-effective deployment of enterprise applications. Secure digital integration ensures seamless communication among heterogeneous systems while maintaining data confidentiality, integrity, and regulatory compliance. This study explores the convergence of these technologies in creating intelligent enterprise ecosystems that support innovation, digital transformation, and competitive advantage. The paper examines existing literature on enterprise intelligence, AI adoption, cloud computing, cybersecurity, and digital integration frameworks. A qualitative research methodology based on secondary data analysis is employed to synthesize current knowledge and identify emerging trends, opportunities, and challenges. The findings indicate that organizations integrating AI with cloud-native platforms and robust security frameworks achieve improved operational performance, enhanced customer experiences, and data-driven strategic capabilities. However, successful implementation requires organizational readiness, skilled personnel, governance mechanisms, ethical AI practices, and continuous cybersecurity management. The study concludes that integrated intelligent enterprise architectures represent the future of sustainable digital business transformation across diverse industrial sectors

Modern enterprises generate and manage enormous volumes of structured, semi-structured, and unstructured information across databases, data warehouses, data lakes, cloud storage systems, enterprise applications, collaboration platforms, and document repositories. Conventional enterprise knowledge-management approaches often struggle to provide timely and contextually relevant access to this distributed information. Retrieval-Augmented Generation (RAG) provides an approach for combining information retrieval with generative artificial intelligence, enabling language models to generate responses based on enterprise-specific knowledge retrieved from authorized data sources. This study examines the integration of modern data-engineering practices with RAG architectures across cloud platforms for enterprise knowledge management. The research focuses on data ingestion, transformation, metadata management, document processing, embedding generation, vector search, retrieval, prompt construction, response generation, and governance. A conceptual methodology is adopted to analyze how cloud-native data pipelines can support reliable and scalable RAG systems while maintaining data quality, security, lineage, and access control. The proposed framework considers heterogeneous enterprise data as a continuously evolving knowledge ecosystem rather than a static repository. It examines the role of batch and streaming pipelines, lakehouse architectures, vector databases, metadata catalogs, and application programming interfaces in establishing an integrated retrieval environment. Particular attention is given to challenges involving data freshness, semantic search accuracy, hallucination, access permissions, privacy, interoperability, and multi-cloud complexity. The study provides a conceptual foundation for implementing enterprise RAG systems through disciplined data engineering and cloud-based infrastructure

Modern enterprise platforms face unprecedented challenges in maintaining operational resilience while managing massive, heterogeneous data streams. This paper explores the integration of Artificial Intelligence (AI) with cloud computing architectures to establish secure, self-healing, and resilient data engineering pipelines. As cyber threats become more sophisticated and data volumes expand exponentially, traditional static infrastructure falls short of meeting enterprise-grade security and availability standards. By leveraging machine learning algorithms, predictive analytics, and automated orchestration within cloud environments, organizations can dynamically mitigate risks, optimize resource allocation, and ensure continuous data integrity. This study evaluates the synergistic relationship between AI and cloud native technologies, detailing how intelligent automation transforms passive data repositories into active, resilient ecosystems. Furthermore, we examine the protocols necessary for securing data engineering pathways, focusing on automated threat detection, zero-trust architectures, and intelligent cryptographic management. Ultimately, this research provides a comprehensive framework for designing next-generation enterprise platforms capable of autonomous adaptation to operational anomalies and security breaches. The findings demonstrate that AI-driven cloud computing not only fortifies data engineering infrastructure against external failures but also significantly enhances compliance, scalability, and long-term operational sustainability in volatile digital landscapes.

The rapid evolution of enterprise digital transformation has intensified the need for intelligent, secure, and automated operational platforms capable of managing complex organizational knowledge across distributed business environments. Traditional enterprise systems often struggle with fragmented data silos, disconnected business processes, limited contextual intelligence, and inefficient decision-making. This research proposes a Knowledge Graph Artificial Intelligence (KGAI)-enabled private cloud automation framework for modernizing enterprise operations through semantic knowledge representation, intelligent reasoning, predictive analytics, and autonomous workflow automation. The proposed framework integrates knowledge graphs, machine learning, deep learning, graph neural networks, cloud-native automation, and private cloud infrastructure to establish a secure and scalable enterprise intelligence ecosystem. Knowledge graphs capture relationships among enterprise entities, business processes, customers, assets, applications, and operational events, enabling AI models to generate contextual insights, optimize workflows, detect anomalies, and support intelligent decision-making. Private cloud environments provide dedicated infrastructure, enhanced security, data privacy, regulatory compliance, and centralized governance while supporting intelligent automation through Kubernetes orchestration, microservices, and Infrastructure as Code. The framework further incorporates Zero Trust Security, explainable AI, identity governance, API security, and continuous monitoring to improve cyber resilience and enterprise governance. By integrating Knowledge Graph AI with private cloud automation, organizations can improve operational efficiency, business intelligence, cybersecurity, predictive decision-making, and digital transformation while maintaining data confidentiality, governance, and long-term enterprise sustainability

The increasing adoption of scalable multi-cloud infrastructures has created complex cybersecurity challenges for modern enterprises. Organizations distribute applications, data, workloads, and services across multiple cloud providers to achieve scalability, availability, flexibility, and cost optimization. However, this distributed architecture expands the attack surface and makes conventional security approaches increasingly difficult to manage. Manual threat investigation and rule-based response mechanisms may be unable to respond rapidly to dynamic and sophisticated cyber threats. This paper proposes an Autonomous Cyber Defense framework based on Reinforcement Learning (RL) for scalable multi-cloud enterprise environments. The proposed framework uses reinforcement learning agents to continuously observe security conditions, evaluate potential threats, select appropriate defensive actions, and learn from the outcomes of previous decisions. The architecture integrates multi-cloud monitoring, identity and access management, anomaly detection, security orchestration, zero-trust principles, automated containment, and cyber-resilience mechanisms. A centralized policy layer coordinates distributed security agents while maintaining cloud-specific enforcement capabilities. The methodology employs a design-oriented research approach involving threat modeling, simulated multi-cloud environments, RL agent development, scenario-based experimentation, and comparative evaluation. Performance is assessed using detection accuracy, response time, false-positive rate, resource utilization, attack containment, recovery time, and policy compliance. The proposed framework aims to improve adaptive threat response, reduce security-operation workload, and strengthen enterprise resilience while maintaining human oversight over high-impact security decisions.

The vision for a cloud-native architecture capable of supporting data exchange and predictive analytics across deployed smart health systems relies on two accepted concepts: (i) interoperability, the ability of an information technology system or product to work with other information technology systems or products; and (ii) predictive analytics, technology that analyzes current and historical facts to make predictions about future events. The proposed architectural framework focuses on data governance, and in particular the aspects of provenance, privacy, and analysis methods. A privacy-preserving approach to data classification detects and replaces sensitive information elements such as social security numbers, addresses, or credit card numbers. The management of data exchange operations is aligned with the requirements of the health domain data model. Predictive analytics use Explainable Artificial Intelligence techniques to provide interpretable predictions for clinical users.

The healthcare domain is one of the sectors where the use of smart systems has been accelerated because of the requirements imposed by the COVID-19 pandemic. Many Cloud-Native Smart Systems are designed, implemented, and deployed on the Cloud to support different Health software applications. Such systems, although deployed in the Cloud, do not have the ability to interact with each other owing to a lack of interoperability. Large amounts of data are collected from different users and devices; the creation of predictive models is an important task to help detect patterns, trends, and predict future values in health, analyzing patients' data to decide the presence of a health problem. A Cloud-Native architecture able to perform predictive analytics in the health domain and that supports interoperability among data is needed.

Cape Verde, an archipelago in the Central Atlantic, is confronted with opportunities and challenges associated with the development of business and tourism sectors, including the root and tuber value chains. This phenomenon is associated with climate change that increasingly dictates new business opportunities and new risk factors to be addressed in the world of tourism. This study identified risk scales associated with climate change along the entire root and tuber value chain in the country. In addition to these factors, 85 positive and negative risk factors were identified, classified and described. A set of thematic maps was built, presenting the risk levels and their factors in Cape Verde and in each of the six islands that make up the country. The research indicated that drought and desertification are the most critical climate risks on the islands. The lower development of the root and tuber tourism value chain in Cape Verde offers new opportunities for businesses and tourism. The analysis of the initialized risk factors in the root and tuber value chain provides the foundation for implementing a climate risk management plan, which seeks to minimize, decrease and/or mitigate the consequences for tourism due to the effects of climate change.

The rapid adoption of artificial intelligence (AI) within enterprise environments has created increasing demand for scalable, reliable, secure, and cost-efficient infrastructure capable of supporting intelligent automation at organizational scale. Cloud computing provides elastic resources, distributed processing, containerized deployment, managed data services, and flexible computational architectures that can support increasingly sophisticated AI workloads. However, deploying AI-driven enterprise solutions introduces substantial infrastructure challenges involving heterogeneous computing resources, large-scale data processing, model serving, workload orchestration, latency management, security, governance, and cost optimization. This study investigates a scalable AI infrastructure framework for next-generation cloud-based enterprise automation and intelligence solutions. The proposed approach integrates cloud-native computing, container orchestration, distributed AI processing, model lifecycle management, scalable data pipelines, observability, intelligent resource allocation, and automated infrastructure management. The research methodology combines systematic literature analysis, architectural modelling, prototype development, controlled experimentation, and quantitative performance evaluation. The framework is evaluated using scalability, latency, throughput, resource utilization, availability, deployment efficiency, model-serving performance, operational cost, and automation effectiveness as key indicators. The study emphasizes elastic infrastructure capable of dynamically matching computational resources to changing AI workloads while maintaining security and service-level objectives. The proposed framework aims to provide enterprises with a flexible foundation for deploying intelligent applications across diverse business functions. The findings are expected to demonstrate how cloud-native AI infrastructure can improve operational agility, reduce infrastructure inefficiencies, and support reliable enterprise-scale intelligence.

Zero-trust cloud operations have emerged as a strategic approach for protecting enterprise environments characterized by distributed workloads, hybrid infrastructure, dynamic identities, and continuously changing attack surfaces. Traditional perimeter-oriented security models are increasingly inadequate because enterprise resources are accessed from multiple locations, devices, applications, and cloud platforms. This research proposes a zero-trust cloud operations framework that integrates autonomous threat detection, continuous security intelligence, behavioral analytics, adaptive access control, and automated response mechanisms to strengthen enterprise resilience. The proposed framework continuously evaluates identity, device posture, workload behavior, network activity, application interactions, and contextual risk before granting or maintaining access. Machine learning models are incorporated to identify anomalous activities and emerging threats, while security intelligence mechanisms correlate telemetry from cloud infrastructure, endpoints, identities, APIs, and applications. A policy-driven zero-trust enforcement layer dynamically adjusts permissions according to changing risk conditions. The research methodology combines architectural design, data preprocessing, machine learning-based anomaly detection, risk scoring, policy evaluation, automated response, and performance validation. Evaluation focuses on threat-detection accuracy, false-positive rate, response latency, policy enforcement effectiveness, resource utilization, and resilience against evolving attacks. The proposed approach demonstrates how autonomous intelligence and continuous verification can transform cloud security from static protection toward adaptive and proactive enterprise defense.

Enterprise cloud infrastructures have become increasingly distributed, dynamic, and complex, creating significant challenges for maintaining service reliability, performance, and operational continuity. Conventional observability approaches primarily depend on retrospective analysis of logs, metrics, traces, and alerts, which can limit the ability of organizations to identify emerging failures before they affect business services. Predictive artificial intelligence (AI)-based cloud observability provides an alternative approach by combining machine learning, statistical forecasting, anomaly detection, and automated correlation techniques to anticipate infrastructure degradation and identify abnormal behavior. This study examines a predictive AI-based cloud observability framework designed to improve enterprise infrastructure resilience and anomaly detection. The proposed framework integrates heterogeneous telemetry from cloud resources, applications, networks, databases, and security systems into a unified analytical pipeline. AI models are applied to establish dynamic behavioral baselines, detect deviations, forecast resource exhaustion, and identify potential service failures. The methodology further incorporates contextual correlation and risk-based prioritization to reduce alert noise and support proactive incident management. Evaluation is proposed using metrics including anomaly-detection precision, recall, F1-score, false-positive rate, forecasting accuracy, mean time to detection, and mean time to recovery. The framework aims to demonstrate how predictive observability can transform cloud operations from reactive monitoring toward proactive resilience management

Enterprise application programming interfaces (APIs) have become essential components of modern digital infrastructures, enabling communication among cloud applications, microservices, business platforms, and third-party systems. However, their widespread adoption has expanded the attack surface, exposing organizations to threats such as unauthorized access, credential abuse, data exfiltration, automated attacks, and API misuse. Traditional security mechanisms based on static rules and perimeter-oriented defenses often struggle to identify sophisticated attacks that resemble legitimate user activity. This research explores an intelligent enterprise API security framework integrating behavioral analytics with adaptive Zero Trust enforcement to enable dynamic threat prevention. The proposed approach examines API traffic patterns, user identities, device attributes, request frequencies, access histories, and contextual risk indicators to establish behavioral baselines and detect deviations. Machine learning techniques support anomaly identification, while continuous authentication, least-privilege authorization, and risk-based access controls regulate suspicious interactions. The framework also incorporates automated policy adaptation, real-time monitoring, and feedback-driven learning to improve detection capabilities against evolving threats. Its effectiveness can be assessed through detection accuracy, false-positive rates, response latency, unauthorized access prevention, and computational overhead. The research proposes a systematic methodology for evaluating security performance under normal and adversarial conditions. By combining behavioral intelligence with continuous trust verification, the framework aims to strengthen API protection, improve threat responsiveness, and support scalable security governance across dynamic enterprise environments

Enterprise cloud workloads exhibit dynamic behavioral patterns due to fluctuating resource utilization, distributed application execution, changing user activities, and automated infrastructure management. Conventional security mechanisms primarily rely on predefined rules and known threat signatures, limiting their effectiveness in identifying previously unseen attacks and subtle deviations from legitimate workload behavior. This study investigates an artificial intelligence-driven framework for behavioral anomaly detection in enterprise cloud environments through continuous profiling and predictive risk assessment. The proposed framework integrates telemetry collection, behavioral feature extraction, baseline profiling, machine learning-based anomaly detection, and contextual risk prediction to identify suspicious activities across cloud workloads. Historical and real-time observations, including CPU utilization, memory consumption, network traffic, authentication events, process activities, and API access patterns, are analyzed to establish behavioral profiles for users, applications, containers, and virtual machines. Anomaly scores are combined with asset criticality, privilege levels, threat intelligence, and deviation severity to prioritize potential security incidents. The research methodology adopts a design-oriented approach supported by controlled experiments and quantitative performance evaluation. Detection precision, recall, F1-score, false-positive rate, detection latency, and predictive risk accuracy are used to assess system effectiveness. The framework also incorporates model explainability, adaptive learning, privacy protection, and human-supervised response mechanisms. The proposed approach aims to improve the early identification of abnormal workload activities, reduce unnecessary security alerts, and strengthen proactive threat management while maintaining the operational reliability and scalability of enterprise cloud infrastructure.