KAKaran Alang

Karan Alang

*This is an algorithmically generated author record based on publicly available information.

Karan Alang has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Optimizing Apache Spark Workflows on Kubernetes for Cloud-Native Environments

    by Karan Alang, Ankush Gumber, Praveen Chaitanya Jakku, Saigurudatta Pamulaparthyvenkata


  • AI-Driven Metadata Extraction and Classification using LLMs in Big Data Lakehouses

    by Bhanuprakash Madupati, Karan Alang, Srikanth Kamatala, Venkata Reddy Pasam, Anil Kumar Jonnalagadda

    Modern data lake houses encounter substantial challenges due to their rapid increase in heterogeneous and unstructured data when it comes to metadata administration and classification and semantic discovery. Such massive and diverse datasets exceed the limits of traditional schema-based methods as well as rigid rule-based systems. This research develops an AI-based method for intelligent classification and automated metadata extraction through Large Language Models. The system uses pre-trained LLMs' contextual understanding and transfer learning abilities to accurately extract metadata from documents logs and multimedia items which have semi-structured and unstructured formats. The framework incorporates a zero-shot learning method from LLMs which teams up with supervised fine-tuning for domainspecific tagging. The validated architecture shows better performance and adaptivity through achieving superior accuracy than traditional NLP-based metadata engines when applied to real-world lakehouse dataset storage. Experimental findings demonstrate that LLMs create substantial enhancements in metadata development as well as search functions and data protection metrics in modern big data systems. This proves LLMs can bring substantial changes to future big data platforms.


  • Governai: Policy-Driven Model Governance for Dynamic and Multi-Tenant AI Systems

    by Sana Zia Hassan, Karan Alang, Nagarjuna Nellutla, Shrinivass Arunachalam Balasubramanian, Vamshidhar Morusu

    Platforms, smart cities, and federated ecosystems, where the governance of models needs to be responsible, and secure, that is, policies-compliant. GovernAI suggests an extended structure that allows intelligent adjustment to the policy-based governance of AI models in the life cycle. GovernAI deals with the main issues of fairness, accountability and compliance management among different stakeholders, by adding declarative policy specifications, tenant-aware access controls, model lineage tracking and real-time behavioral audits. This architecture is capable of feature dynamic policy enforcement, automatic model versioning and dispute resolution in shared landscapes. Experimental analyses prove that GovernAI works in real-world situations minimizing the number of policy violation by 45 %, improves auditability in the multi-tenant setting and allows timely action to counter model misbehavior or adversarial attacks. Touching on ethical AI principles, this work is an application of the domain-specific governance layer that makes ethics scalable and interpretable across the whole AI domain and down to the most complex AI infrastructures.


  • Autonomous Data Quality Monitoring with AI Agents: Integrating ML with Cloud Warehouses and Data Lakes

    by Srinivas Chippagiri, Karan Alang, Ankush Gumber, Sooraj George Thomas

    Since storing data in cloud warehouses and lakes has increased, it is now important to rely on automation for keeping data accurate and trustworthy. It outlines an autonomous way of monitoring data quality by using AI agents with botnet techniques and cloud architecture. Key features of the system are that it constantly scans the data, finds unusual activities, detects any changes in the structure, and responds to any data updates without someone needing to act. Missed values, duplication, inconsistency, and noticing outliers on both structured and semi-structured data is done by using both supervised and unsupervised ML models. AI agents interact by way of a network pipeline and use reinforcement learning to pick the best solutions. easily set up Cloud Genomics in any leading cloud environment, and it works well with data warehouses as well as data lakes. Experimental evidence shows that AI-driven analysis is better at detecting errors, offering faster results, and being more adaptable than hand-made and rule-based ones. In this research, data pipelines can become more flexible and easier to manage in today's business architecture.


  • Real-Time ML and LLM Optimization: Orchestrating Scalable Workflows in Distributed Commerce Environments

    by Karan Alang, Sana Zia Hassan, Venugopal Katkam, Shazia Hassan

    Due to the exponential increase of the data in the distributed commerce settings, new issues of the real-time decision making and operational scale have emerged. In order to solve these problems, this paper proposes an orchestration framework that brings together real-time machine learning (ML) models with large language models (LLMs) to optimize distributed workflows in e-commerce ecosystems. The suggested framework relies on micro services-based architecture, edgecloud synergy, and semantic data pipelines to enable low-latency inference, adaptive learning, and context-aware automation. With the integration of reinforcement learning to adapt dynamic workflows and intelligent task coordination and personalized customer interactions based on LLMs, the system provides efficiency in operations and enhanced user interaction. Evaluations on a real-world retail data set show, under varying load conditions, large latency reduction and throughput gains as well as high model accuracy. The scalability, fault tolerance and DataOps/MLOps pipeline integration are also studied, which makes the model applicable to enterprise-level deployment across multi-cloud setups.

  • Open Access

    Machine Learning for Anomaly Detection and Prediction in Network Data

    by Karan Singh Alang, Shantanu Bindewari

    As network infrastructures expand in scale and complexity, the ability to detect anomalies and predict network disruptions has become a critical component of cybersecurity and operational reliability. This research investigates the application of machine learning techniques to improve anomaly detection and forecasting in network data. By leveraging a combination of supervised and unsupervised algorithms, our framework integrates clustering methods, support vector machines, and deep neural networks to identify irregular traffic patterns and anticipate potential failures. Experimental evaluations on multiple real-world datasets demonstrate that the proposed approach outperforms traditional rule-based systems in terms of detection accuracy and false-positive reduction. The model’s capacity to adapt to evolving network behaviors and its scalability in processing large volumes of data provide significant advantages for proactive network management. In addressing challenges such as feature selection, data imbalance, and computational overhead, our work emphasizes the importance of algorithmic tuning and robust preprocessing techniques. The predictive analytics component further enables network administrators to implement timely interventions, reducing downtime and mitigating risks associated with cyber threats. Overall, our findings suggest that integrating advanced machine learning methods into network monitoring systems offers a viable and effective solution for maintaining high levels of security and operational performance. This study lays the groundwork for future research aimed at enhancing anomaly detection and predictive capabilities in dynamic network environments while promoting a safer digital infrastructure. The promising results achieved underscore the potential of machine learning to revolutionize network security protocols, ensuring responses to emerging threats and reinforcing the stability of communication infrastructures.


  • Stream Processing with Apache Kafka: Real-Time Data Pipelines

    by Andhra University Alumnus, Karan Singh Alang, Ajay Shriram Kushwaha, Sharda University

    Apache Kafka has emerged as a pivotal technology in the realm of real-time data processing, enabling the construction of robust and scalable stream processing systems. This paper explores the utilization of Apache Kafka as a backbone for real-time data pipelines, detailing its capacity to ingest, buffer, and process continuous streams of data with high throughput and minimal latency. Kafka’s architecture is built on a distributed, fault-tolerant model that leverages partitioned logs to ensure data consistency and availability even amid node failures. By employing a producer–consumer paradigm, organizations can handle vast volumes of data concurrently, thereby supporting rapid insights and timely decision-making. The integration of Kafka with complementary frameworks—such as Kafka Streams, Apache Spark, and Apache Flink—further enhances its ability to perform complex event processing, aggregation, and transformation tasks. Real-world applications in industries like finance, telecommunications, and IoT underscore the platform’s versatility in meeting the demands of modern data-driven operations. This paper also examines challenges including data consistency management, backpressure mitigation, and the seamless scaling of infrastructure. Through detailed analyses and case studies, the discussion illustrates how Apache Kafka not only addresses the limitations of traditional batch processing but also lays a flexible foundation for next-generation analytics platforms. Overall, this exploration provides valuable insights into harnessing Kafka’s full potential in the era of real-time digital transformation, affirming its role as an indispensable component in contemporary data engineering.