Hrushikesh Chavan

Leading generative AI, Agentic AI and LLM projects, including a RAG-based chatbot and solutions using GCP Vertex AI, Azure AI, Azure Cognitive Services, Azure OCR, Copilot Studio, GPT 4o and Claude 3.5, with a strong focus on healthcare transformation. Managing a team of 7 to drive project delivery and ensure alignment with healthcare objectives. Designed generative AI chatbots to enhance customer engagement and streamline support, and developed agentic AI POCs using frameworks like Crew AI and AutoGen.

Generative AI & ML Engineer. Built chatbot with sentence transformers, LangChain, and Azure OpenAI GPT-3.5 for efficient document retrieval. Developed video generation system using Multimodal RAG, LlamaIndex, and text-to-speech synthesis. Automated product reliability analysis using survival analysis and PySpark, achieving 83% runtime reduction and $65K cost savings. Implemented ML-based defect prediction to reduce development costs. Led 20-member team fostering innovation and collaboration

Led cross-functional data science initiatives and a 5-member team for on-time delivery. Built NLP solutions (text classification, summarization, NER, sentiment analysis) using PyTorch, TensorFlow, and Hugging Face (BERT, DistilBERT, T5, PEGASUS). Developed aerospace ML models: predicted hose fitting failures via DOE simulations (60% faster data acquisition), improved enquiry conversions from 30% to 70% using end-to-end ML deployment.

Developed Open-Source Python-Based Tool for Generating STL Files of TPMS Surfaces for 3D Printing • Developed a Python-based tool that enhanced the generation of complex 3D printed structures by applying advanced algorithmic techniques and leveraging open-source libraries. This significantly improved the efficiency and practicality of 3D printing, and saved license costs associated with proprietary software. Python, Numpy, Scikit-Image, Meshgrid, STL

Conducted insightful exploratory data analysis, discerning trends, patterns, and correlations to cater to the diverse analytical needs Implemented machine learning systems for NBFC client: Credit Card Approval System: Attained a notable 90% accuracy in evaluating creditworthiness and detecting potentially fraudulent activities Loan Approval System: Demonstrated a significant 15% improvement in loan processing efficiency