My Research Area

My work centers on designing intelligent systems that combine GenAI, multi-agent architectures, cloud-scale decision intelligence, and AI-native network principles. These four pillars define the core of my technical and research interests.

1. GenAI, LLM Systems & Reasoning Architectures

I focus on developing GenAI and LLM-driven systems that enable reasoning, planning, and context-aware decision-making. Key areas include:

  • LLM orchestration and multi-step reasoning

  • Retrieval-Augmented Generation (RAG) and vector search

  • Tool-using LLM systems and function-calling pipelines
  • Hallucination mitigation and reliability strategies
  • Evaluation frameworks for reasoning quality and performance autonomy
      • My goal is to design GenAI systems that integrate reliably into enterprise and network environments, where structured decision-making and system-level consistency are essential.

      2. Multi-Agent Systems & Agentic Intelligence

      I study architectures where multiple agents collaborate, coordinate, and act autonomously. My work includes:

    • Multi-agent orchestration and task delegation

    • Inter-agent communication and coordination mechanisms

    • Autonomy levels and safety guardrails

    • Policy-driven agent behavior

    • Integration of LLM reasoning into agent workflows

      • This pillar focuses on enabling systems where agents are intelligent, observable, aligned, and capable of autonomous action.

      3. AI-Native Networks & Distributed Intelligence (5G/6G)

      With a background in advanced 5G/6G architecture, I explore how AI reshapes network design. Areas include:

    •  Intelligent network functions (AMF, SMF, PCF, NWDAF)

    •  AI capability exposure and cross-domain intelligence

    •  QoD (Quality of Decision) and metadata for decision-making

    •  Autonomous mobility, roaming, and capability negotiation

    •  Distributed learning and device-level intelligence

      • This work bridges networking and AI, defining architectures that are AI-native, data-driven, and autonomous — essential foundations for 6G

      4. Cloud AI Systems & Data/QoD Intelligence

      I design cloud-scale architecture that supports complex AI workloads, with a focus on:

    •  Cloud-native AI pipelines (AWS, Oracle, hybrid environments)

    •  Vector databases and semantic retrieval

    •  calable inference, agent layers, and orchestration

    •  Metadata-driven decision frameworks (QoD, observability, governance)

    • Data pipelines, telemetry, and AI system monitoring

      • This pillar connects enterprise AI deployment with robust, intelligent, production-ready infrastructure.