Neeraj Kumar Gontala Montgomery

Designed and deployed an enterprise Financial Intelligence Assistant using Retrieval-Augmented Generation (RAG), LangChain, Hugging Face Transformers, and vector search to improve response accuracy to 94% while reducing customer query resolution time by 65%. Built production conversational AI for customer support, financial document search, and knowledge retrieval supporting 50,000+ monthly users. Engineered semantic search pipelines with transformer embeddings and optimized retrieval for sub-second access to policies, regulatory documents, contracts, and enterprise knowledge. Developed agentic AI workflows integrating tool calling, function orchestration, retrieval pipelines, and enterprise APIs to automate financial research and decision support. Collaborated across engineering and product teams to deploy AI into enterprise applications and establish model evaluation frameworks to improve scalability, reliability, and user experience.

Developed machine learning models using XGBoost, LightGBM, and scikit-learn for pricing optimization, customer segmentation, and predictive analytics, improving pricing accuracy and increasing revenue by 18%. Built and deployed end-to-end AI pipelines on AWS and Azure using Docker, MLflow, Jenkins, and FastAPI, reducing deployment cycles by 35% and enabling automated retraining and monitoring. Created recommendation engines from customer transaction history and behavioral analytics to increase product adoption and customer engagement. Implemented RAG applications with LangChain and transformer embeddings to automate document retrieval, knowledge discovery, and workflow automation, improving relevance and reducing irrelevant results by 22%. Built deep learning models in PyTorch and TensorFlow for computer vision, NLP, and speech tasks with accuracy above 92%. Established MLOps workflows with MLflow for experiment tracking, model versioning, explainability, lineage, and drift monitoring,