Hruthik Gonuguntla

Designed and implemented enterprise-scale AI/ML and Generative AI solutions leveraging RAG, MCP-enabled architectures, agentic AI, LLMs, NLP, and Transformers in AWS environments. Architected end-to-end RAG pipelines with ingestion, chunking, metadata extraction, embeddings, vector indexing, semantic retrieval, reranking, contextual prompting, and grounded response generation. Built AI agents and chatbots using multi-step reasoning, tool calling/function invocation, planning, memory, and orchestration patterns. Developed NLP pipelines for healthcare text preprocessing and analysis, and built scalable semantic search and knowledge-retrieval architectures using embeddings and vector databases (e.g., FAISS). Implemented GenAI automation for summarization, extraction, classification, QA, and knowledge discovery, and optimized inference and retrieval quality with evaluation frameworks. Exposed ML/LLM capabilities through secure REST APIs (FastAPI/Flask), integrated with enterprise systems,

Designed and implemented enterprise-scale AI/ML and Generative AI solutions leveraging RAG, MCP-enabled architectures, agentic AI, NLP/Transformers, and LLMs on AWS. Built end-to-end RAG pipelines for ingestion, chunking, metadata extraction, embeddings, vector indexing, retrieval, reranking, contextual prompting, and grounded response generation. Developed tool-using AI agents for multi-step workflows, planning, memory, and orchestration with secure tool calling. Created member-facing conversational AI and chatbot backends via REST services. Engineered semantic search and knowledge-retrieval systems using embeddings, FAISS/vector DBs, and evaluation frameworks to reduce hallucinations and improve answer relevance. Built LLM-powered document intelligence workflows (summarization, extraction, classification, QA) and implemented model optimization and systematic evaluation using precision/recall, F1, AUC-ROC, and retrieval/grounding metrics. Delivered scalable Python APIs with FastAPI/Fl

Developed enterprise AI and ML solutions on Azure combining NLP, LLMs, intelligent search, conversational AI, deep learning, and predictive modeling. Built chatbots and knowledge-search applications using intent classification, entity extraction, dialogue management, contextual embeddings, and retrieval-based techniques. Designed intelligent search and navigation strategies using heuristic and pathfinding algorithms (A*, BFS, DFS, etc.) alongside semantic similarity and ranking. Built Transformer-based NLP pipelines for classification, semantic similarity, question answering, and summarization. Implemented embedding and semantic-search pipelines using vector indexing for context-aware retrieval across large unstructured datasets. Delivered ML-ready datasets using PySpark/SQL, deployed model-serving APIs with FastAPI/Flask, and improved reliability through systematic evaluation (AUC-ROC, precision/recall, F1, cross-validation).