Developed a multi-agent AI architecture consisting of a Main Orchestrator Agent and specialized sub-agents (Flight, Card,Lounge Management) to handle complex airport lounge workflows.
• Implemented LangChain-based agent orchestration, enabling the main agent to dynamically route user queries, invoke relevant sub-agents, and aggregate responses into a single coherent output.
• Built a Flask-based REST API to expose AI agent functionalities, supporting real-time user queries and seamless backend–agent
communication.

Authored detailed reports evaluating multiple LLM models on identical prompts, analyzing response quality, reasoning
behavior, and instruction adherence.
• Documented internal working and real-world applications of Generative AI, focusing on enterprise automation and real-time
decision support.
• Concluded from LLM benchmarking that ChatGPT is better suited for general reasoning, while Grok performs better on
coding-focused workloads.