I am currently engaged in in-depth research and applied exploration of Model Context Protocol (MCP), Agent-to-Agent (A2A) Protocols and agentic AI architectures, with a specific focus on their implications for edge computing and distributed systems.
This work sits at the intersection of artificial intelligence, systems architecture, and real-world infrastructure, where scalability, latency, resilience and autonomy are critical constraints rather than abstract concepts.
My research examines how modern AI agents can move beyond centralized cloud-bound models and operate effectively at the edge—closer to data sources, devices, and users.
By leveraging agentic architectures, I am exploring how autonomous agents can reason, coordinate, and act across heterogeneous environments, even under limited connectivity or compute resources. Protocol-level design is a core area of interest: MCP enables structured context sharing and tool interoperability, while A2A protocols introduce standardized communication, negotiation and collaboration patterns between intelligent agents. Together, these paradigms offer a foundation for scalable, composable, and vendor-agnostic AI systems.
A key component of this work involves exploring the capabilities of open-weight LLMs for network-constrained environments. This approach opens the door to AI-driven systems that can manage infrastructure complexity, optimize resource allocation, monitor anomalies, and respond dynamically to changing conditions in a decentralized manner, without constant human intervention.
My goal is to translate these emerging concepts into practical edge-native solutions — from intelligent monitoring and orchestration layers to self-healing infrastructure and decentralized decision-making systems. By combining protocol-level rigor with agent-based intelligence, I aim to design architectures that are not only powerful, but also transparent, modular, and resilient. This research reflects a broader commitment to building AI systems that solve meaningful, real-world problems, particularly in environments where reliability, efficiency, and autonomy are paramount.