Bridging Classroom Learning to the real World through Emerging Computing Paradigms

The School of Business & Management – BBA, CHRIST (Deemed to be University), Pune Lavasa Campus, organized an international webinar titled “Bridging Classroom Learning to the Real World Through Emerging Computing Paradigms” on 16th January 2026, from 10:30 AM to 11:20 AM. The session was conducted in collaboration with Multimedia University (MMU), Malaysia, under the framework of the Memorandum of Understanding (MoU) signed between the two institutions, highlighting the commitment to global academic collaboration and knowledge exchange. The webinar was delivered through the Google Meet platform and was attended by first-year BBA students.

The session was led by Prof. Ts. Dr. R. Kanesaraj Ramaswamy, Associate Professor, Faculty of Computing and Informatics, Multimedia University, Malaysia. The primary objective of the webinar was to bridge the gap between theoretical classroom learning and real-world application by exposing students to emerging computing paradigms, particularly Artificial Intelligence (AI), Machine Learning (ML), and advanced computing systems used in industry.

Dr. Ramaswamy emphasized that while classroom learning often focuses on achieving high accuracy in controlled datasets, real-world systems are far more complex and uncertain. He explained that even models with 98% accuracy can fail in production environments due to issues such as class imbalance, rare events, fraud cases, and data misconceptions. The session highlighted critical challenges faced in real-world deployments, including domain shift, where models trained in one geographical or operational context may fail in another due to sensor variations or environmental differences, and concept drift, where patterns in data change over time.

A key learning from the session was the distinction between an ML model and an ML pipeline. While students typically submit trained models as part of academic assessments, industries deploy comprehensive ML pipelines that include multiple stages such as data validation, feature engineering, model training, inference APIs, monitoring, and drift detection. The speaker stressed that AI is not just about model training but about building robust systems that can adapt to real-world uncertainties.

The discussion also covered overfitting to clean or artificial data, explaining that models performing well on ideal datasets may become operationally ineffective in real-world scenarios. Dr. Ramaswamy elaborated on failure modes in deployed models, such as how Convolutional Neural Networks (CNNs) may experience confidence collapse under noisy conditions, and compared them with newer architectures like Transformers. The session further explored model robustness and explainability, underlining the need for benchmarks, controlled noise levels, and transparent decision-making processes to ensure trust and reliability. Emerging paradigms such as Edge and Cloud computing were discussed, highlighting their applications, advantages, and limitations. While edge computing enables real-time processing closer to data sources, cloud computing offers scalability and high data-handling capacity. Challenges such as edge failures due to network unavailability and low-confidence scenarios were also addressed.

Towards the end, the speaker introduced students to Quantum Computing, discussing its potential, limitations, high cost, and applications in cryptography and pharmaceutical research. He concluded by emphasizing that while paradigms evolve, fundamentals persist, and real-world systems must always be rigorously tested. Overall, the session effectively connected classroom concepts with industry realities and reinforced the importance of interdisciplinary learning. As part of the MoU between CHRIST (Deemed to be University) and MMU, Malaysia, the webinar added significant value by providing students with global perspectives on emerging technologies and their real-world applications.