Jay Parmar serves as Lead AI Systems Architect at Healioscan, where he spearheads the design and deployment of next-generation AI infrastructure that powers real-time diagnostics and predictive health intelligence. With over 5 years of hands-on leadership in Python, large language models, and cloud-native AI, Jay has architected scalable, production-grade systems on AWS—leveraging SageMaker, Lambda, S3, and serverless frameworks—to deliver mission-critical solutions at enterprise velocity. A pioneer in multi-agent RAG architectures, Jay engineered Healioscan’s flagship intelligent assistant using LangGraph and LangChain, enabling seamless natural-language interaction with structured SQL databases and unstructured clinical documents. This system not only answers complex queries with sub-second latency but autonomously creates and updates patient profiles, slashing HR overhead by 70%+ in pilot deployments. His NER-driven document automation pipeline eliminates manual contract review, processing thousands of pages daily with near-perfect accuracy. Previously, as Python Team Leader at Thinkbiz Technology, Jay directed cross-functional squads—collaborating with Scrum Masters, PMs, and QA—to deliver generative AI innovations that consistently outperformed benchmarks in contextual relevance and efficiency. His deep expertise in prompt engineering, vector databases, and fine-tuning has made him a go-to authority for transforming raw LLMs into domain-specific powerhouses. A Certified AWS AI Practitioner with an Masters of Technolgy in Computer Science from Nirma University, Jay’s research has been published in leading journals, including groundbreaking work on microservice observability and real-time face mask detection using MobileNetV2. His technical arsenal—spanning Deep Learning, DevOps, Docker, and microservices—positions him at the forefront of AI systems engineering. At Healioscan, Jay isn’t just building models—he’s orchestrating intelligent ecosystems that redefine healthcare automation.