Dr Johan Barthelemy1, Dr Wei Fang1, Bharatkumar Sharma1
1NVIDIA, , Australia
Biography:
After completing his PhD in Applied Mathematics at the University of Namur in Belgium, Dr. Johan Barthélemy joined the SMART Infrastructure Facility at the University of Wollongong in Australia. There, he led the Digital Living Lab, focusing on the research and development of AI and AIoT solutions for smart cities applications and biodiversity monitoring in extreme environments. Passionate about applied AI and accelerated computing, he joined NVIDIA in 2022 where he is now a Senior AI Scientist at the NVIDIA AI Technology Center, where he collaborates with developers and researchers in building the next generation of AI-powered solutions.
Dr. Wei Fang is a Senior Solutions Architect at NVIDIA, helping researchers across Australia and New Zealand harness AI and accelerated computing to advance scientific discovery. He works closely with the research community to deploy and scale cutting-edge solutions using NVIDIA’s full-stack platform, spanning hardware, software, AI models, and application frameworks.
Bharatkumar Sharma has 10+ years of development and research experience in the domain of Software Architecture, Distributed and Parallel Computing. He has published papers, journal articles and books in field of Parallel Computing and Software Architecture. He is currently heading the NVIDIA AI Technology Center (NVAITC) India operations. He has expertise in designing Software for Non Functional Requirements and Software Migration to latest heterogeneous parallel architectures. Bharat received his master degree in Information Technology from Indian Institute of Information Technology (IIIT, Bangalore)
Abstract:
As Generative AI evolves, research institutions are transitioning from passive chat interfaces to autonomous, goal-driven AI agents capable of reasoning, planning, and executing complex tasks. This hands-on workshop introduces participants to the fundamentals of Agentic AI using NVIDIA’s open-source framework, including NeMoClaw and OpenShell.
Designed for research software engineers, data specialists, and IT infrastructure professionals, this session moves beyond traditional prompt engineering to explore how autonomous agents can significantly enhance research productivity and data workflows.
Participants will learn how to design, implement, and orchestrate specialised agents or “claws” that function as persistent, background research assistants. The session will cover the following topics:
– Construct a functional autonomous agent tailored for domain-specific research applications.
– Master architectural blueprints and reusable patterns that facilitate rapid team integration.
– Mitigate technical debt by validating infrastructure choices against industry-standard frameworks.
The local execution models provide flexible frameworks and attendees will leave equipped with the practical knowledge needed to build secure, domain-specific AI assistants. This workshop bridges the gap between state-of-the-art AI tooling and daily research data infrastructure, empowering institutions to shape the future of automated, reproducible research.
