Professional profile, for research and industry contacts.
Trace-driven, role-aware benchmark for evaluating LLM agents in HPC operational contexts under role-based access-control policies.
Modular LLM-orchestrated agent framework for end-to-end Kubernetes management, with human-in-the-loop approval on every mutating operation. Published in the Journal of Grid Computing (2026).
Open-source software released alongside the research: agent frameworks, benchmarks and tooling for HPC and Kubernetes.
Full list of publications and citation metrics.
Personal research website: projects, open-source software, publications and contact details.
