Jisvitha Athaluri

AI/ML Engineer with 5 years of experience building and deploying production machine learning systems across insurance, mobility, and industrial domains using Python, PyTorch, SQL, and Spark. Built scalable real-time and batch inference systems supporting high- volume enterprise workloads. Designed and operationalized Retrieval-Augmented Generation (RAG) pipelines using LangChain, Pinecone, LLAMA, improving knowledge retrieval precision by 27% across 500K+ documents and accelerating customer support response cycles by 38%. Implemented Bayesian hyperparameter optimization, cross-validation, and drift detection frameworks to improve prediction performance by 11% and reduce degraded model exposure by 25% in production environments. Deployed model serving infrastructure with FastAPI, Docker, Kubernetes, and MLflow, enabling sub-2-second inference latency for 500+ concurrent requests. Strengthened ML lifecycle governance through experiment tracking, A/B testing, and model monitoring, ensuring reproducibility and SLA compliance across large-scale production environments.