Gnaneswar Goddu

• Designed scalable ML models for sales insights and risk scoring, improving forecasting accuracy by 23% and reducing manual reporting effort by 40%. • Built NLP pipelines with Llama 3 & BERT, processing 50K+ daily customer conversations, boosting classification accuracy by 18% for insurance client support. • Designed and developed data warehousing solutions, including ETL processes to load enterprise data into the Data Warehouse, applying dimensional modeling techniques for optimized data retrieval. • Developed standardized ETL pipelines in Azure Data Factory, integrating 10+ disparate data sources, cutting data prep time by 35%. • Integrated LLM-driven NLP models into automation workflows, enabling real-time sentiment and intent detection, reducing manual intervention in customer support processes by 40%. • Implemented retrieval-augmented generation (RAG) workflows with vector databases to provide domain-specific responses, increasing knowledge query accuracy by 25%. • Optimized LLM inference pipelines by deploying quantized models in Azure ML and Databricks, cutting inference latency by 35% while maintaining accuracy. • Established prompt engineering best practices for insurance and financial services use cases, boosting LLM response relevance and reducing error rates by 20%.