Edison Kagona is a trailblazing expert at the nexus of artificial intelligence, software engineering, and education. With over a decade of diverse experience, Edison is dedicated to harnessing AI and digital technologies to revolutionize quality education, foster innovation, and address systemic challenges in Africa and beyond.
As Senior Tech Lead at Turing Inc., based in Silicon Valley, Edison designs and implements cutting-edge, scalable AI- driven applications. His work spans the full spectrum of software development, data engineering, machine learning, and AI systems, where he leads cross-functional teams to create intelligent solutions—ranging from predictive analytics to natural language processing—addressing real-world problems with practical, innovative applications. His portfolio reflects a relentless pursuit of excellence in deploying robust and adaptable AI systems that catalyze digital transformation.
In Uganda, Edison serves as the Manager of the Centre for Innovation and Entrepreneurship at Cavendish University Uganda, where he spearheads initiatives that nurture entrepreneurial mindsets and digital literacy among students and local communities. His leadership facilitates the development of practical, AI-enabled projects that empower the next generation of innovators.
Edison’s experience extends to impactful research and development roles, such as Senior Data Scientist and Software Technology Lead at M-Omulimisa (Agritech Company), where he developed predictive models to support agricultural decision-making projects funded by GIZ and Mastercard, Lecturer and Research Fellow at the International University of East Africa, Cavendish University, Nkumba University and Victoria University where he contributed to academic excellence by delivering courses, supervising research, and publishing on AI, blockchain, and e-learning systems. His publications include influential articles on AI-driven exam clearance, blockchain-based voting protocols, and augmented reality in e-learning.
A recognized thought leader, Edison has presented and moderated at prominent international conferences, including the Nkumba International Research Conference 2021 and the African Data Engineering and Science Conference (ADESA). His upcoming paper, selected for the ACM / IEEE Publication and ICSE 2025 (47 th International Conference on Software Engineering) in Ottawa, Ontario, Canada, critically analyzes the policy, legal, and technical barriers to the commercialization of AI and IoT-driven agricultural solutions in the East African Community—a testament to his commitment to advancing AI adoption in critical sectors.
Edison holds a master’s degree in computer security and a bachelor’s in information technology, complemented by certifications in network security and data engineering. His professional journey reflects a fusion of academic rigor, innovative leadership, and practical impact. As a passionate advocate for AI’s role in democratizing quality education and empowering communities, Edison embodies the vision of a future where technology bridges gap, fosters inclusivity, and drives sustainable development.
Intelligent, Pre-Deployment Verification of Smart Contracts via Decentralized Machine Learning
The transformative potential of blockchain and smart contracts is critically tempered by the persistent and economically significant threat of exploitable vulnerabilities. Existing smart contract verification methodologies, often reliant on computationally intensive on-chain analysis or fallible manual audits, struggle to provide the necessary proactive security guarantees demanded by high-stakes decentralized applications.
This research pioneers a fundamentally novel approach: the synergistic integration of blockchain technology with decentralized machine learning deployed at the network edge. Our architecture introduces "Sentinel Nodes", edge computing devices imbued with rigorously trained machine learning models capable of performing sophisticated, pre-deployment analysis of smart contract code and token characteristics. This distributed intelligence framework enables real-time identification of subtle vulnerability patterns and comprehensive token integrity checks with minimal latency overhead on the core blockchain. By shifting the verification paradigm to the network's periphery, we not only enhance the proactive security posture of smart contracts but also address critical scalability limitations inherent in centralized verification processes.
Through a rigorous methodology encompassing theoretical design, practical implementation on a representative edge computing platform, and comprehensive empirical evaluation against known vulnerability datasets and real-world smart contracts, this research demonstrates a significant leap forward in smart contract security and efficiency. The findings underscore the efficacy of decentralized machine learning as a robust, scalable, and intelligent solution for ensuring trustworthiness and widespread adoption of blockchain-based decentralized applications.