▪ Architected and deployed scalable LLM serving infrastructure utilizing Neural Information Manager (NIM), supporting efficient, high-throughput model inference and seamless resource management and load balancing across distributed multi-GPU setups.▪ Architected the design and development of a Knowledge-Bot with Agentic AI using Crew AI Agentic framework, transforming it into an autonomous agent capable of making its own decisions and taking actions to achieve specific goals.▪ Led the design and implementation of state-of-the-art Generative AI models using Large Language Models (LLM) at scale, enabling clients to generate high-quality, context-aware content and automate complex tasks like Insurance Underwriting, Knowledge Bot etc.▪ As part of the core senior technology Innovation leadership team, trained, developedand coached managers and individual contributors on Generative AI technical andprofessional skills.▪ Led the design, development and execution of Generative AI strategies, leveraging LLMs for various applications in insurance domain.▪ Optimized LLM inference pipelines by integrating frameworks like TensorRT and vLLM in multi-tenant LLM serving architectures with dynamic scaling, supporting varying workload demands, achieving significant reductions in latency and resource consumption for real-time applications, thus enabling reliable and cost-efficient service delivery.▪ ⁠Fine-tuned SOTA large language models like llama 3 and 3.1 at scale using FSDP and Deep Speed in AWS SageMaker to adapt to insurance domain for Underwriting Application.▪ Provided technical expertise and strategic guidance to executive leadership on the adoption and deployment of Generative AI technologies within the business.▪ Led Data Science teams, collaborated with business analysts, developers and technical support teams to define project requirements and specifications and translating them to project scope of work.