Mohit Bajpai
Advancing Network Intelligence with Generative AI: From Anomaly Detection to Autonomous Remediation
Abstract:
Modern digital infrastructures—spanning cloud computing, edge environments, and 5G networks—demand intelligent monitoring systems capable of identifying anomalies and predicting failures before they disrupt services. Traditional rule-based monitoring frameworks rely heavily on static thresholds and signature-based detection, making them ineffective in detecting emerging threats and dynamic network behaviors.
In this keynote, Mohit Bajpai presents a Generative AI–driven framework for proactive network performance monitoring, anomaly detection, and autonomous remediation. The proposed approach leverages advanced generative models—including Variational Autoencoders (VAE), Generative Adversarial Networks (GAN), and TimeGAN—to learn complex patterns from network telemetry data and detect deviations indicative of faults or cyber threats. By generating synthetic network scenarios, these models enhance the detection of rare and zero-day anomalies that conventional methods often miss.
The architecture integrates LLM-based reasoning and reinforcement learning to recommend or automatically execute corrective actions through software-defined orchestration systems, enabling self-healing network capabilities. Experimental evaluations across multiple datasets demonstrate improved anomaly detection accuracy, reduced false positive rates, and faster mean time to detect compared with traditional machine learning techniques.
The session concludes by discussing future directions for AI-powered autonomous networks, highlighting how generative intelligence can transform network operations, enhance resilience, and support the next generation of intelligent digital infrastructure
Profile:
Mohit Bajpai is a technology leader and researcher with over 20 years of experience in software engineering, network monitoring, cloud computing, and intelligent automation. He currently serves as Senior Software Engineer III at Softility Inc., where he develops scalable solutions for large-scale infrastructure environments using cloud platforms, automation frameworks, and AIOps technologies.
His expertise spans AI-driven network monitoring, DevSecOps, cloud-native architectures, and telecom systems, with extensive experience implementing monitoring,troubleshooting and event management solutions. His work focuses on building predictive, automated, and self-healing systems that improve operational efficiency and system reliability.
Mohit has authored multiple research publications on network automation, AIOps, and cloud-based monitoring systems, contributing to the advancement of intelligent infrastructure management. He is a Senior Member of IEEE and serves as a reviewer for several international journals. His work has also received recognition through multiple innovation awards and international research honors.
Through his research and industry contributions, Mohit continues to explore the intersection of Generative AI, predictive analytics, and autonomous network operations to drive the next generation of intelligent digital infrastructure. His expertise in building predictive, self-healing systems and intelligent monitoring frameworks will contribute practical perspectives on advancing resilient, autonomous network operations for next-generation digital ecosystems.