Shubham Ragade

I’m a Data Scientist and Generative AI Engineer with 2+ years of experience building enterprise-scale AI, ML, Deep Learning, and LLM-powered applications. I specialize in RAG, Fine-Tuning, Prompt Engineering, AI Agents, NLP, and scalable MLOps pipelines.

At Smart Software Services Pvt Ltd., I architected LLM-based Generative AI applications using LangChain, RAG pipelines, and Vector Databases. I designed and deployed AI chatbots with OpenAI APIs, Pinecone, and FastAPI—reducing manual customer support workload by 45% and average response time by 35%. I also fine-tuned transformer-based LLMs using LoRA/QLoRA, improving response relevance by 30% while reducing fine-tuning cost.

I focus on production outcomes: I built end-to-end MLOps pipelines with MLflow, Docker, and AWS SageMaker to enable zero-downtime deployment, automated versioning, and real-time monitoring. I optimized ETL and real-time workflows using Apache Airflow, cutting processing time by 50% and improving reliability to 99% uptime.

Previously, as a Generative AI & Data Science Intern, I developed LLM + RAG solutions for semantic search and automated summarization, reducing manual review time by 40%. I bring that same hands-on engineering mindset to my projects—fraud detection with 95% precision and AUC 0.91, and agentic customer support automation that handled 70% of queries without human intervention.