Sai Prakash Reddy Adudotla

Ai Ml Engineer Gen Ai Llm Specialist Agentic Ai Developer @AdventHealth

US

Led the design, development, and deployment of Generative AI systems using GPT-3.5, GPT-4, and fine-tuned LLMs tailored for healthcare and insurance applications such as claim analysis, document summarization, and intelligent triage. • Built agent-based AI frameworks using LangChain, AutoGPT, and BabyAGI to execute complex tasks autonomously, including knowledge retrieval, clinical analytics workflows, and dynamic multi-step decision-making. • Engineered healthcare-optimized LLMs with Reinforcement Learning with Human Feedback (RLHF), improving capabilities in summarization, retrieval, claim processing, and compliance-driven document interpretation. • Constructed retrieval-augmented generation (RAG) systems using AWS Kendra, Pinecone, Weaviate, and FAISS to enhance LLM response relevance through semantic indexing and ranking strategies. • Applied Hugging Face Transformers, LoRA, and QLoRA techniques to fine-tune LLMs on large insurance/medical corpora with reduced memory usage, enabling efficient training at scale. • Designed robust MLOps pipelines across GCP Vertex AI, AWS SageMaker, and Azure ML, incorporating CI/CD, monitoring, and automated deployment of LLM applications. • Created Transformer-based ICD/DRG prediction models and time-series forecasting pipelines using BERT, TensorFlow, and Keras, improving risk prediction accuracy and cost trend monitoring • Engineered ingestion pipelines with Kafka, Beam, and Dataflow for structured/unstructured data, built knowledge graphs with Neo4j/networkx, and migrated legacy systems to BigQuery/Redshift.