Sr Generative AI Engineer at AT&T

Designed, developed, and deployed production-grade generative AI solutions for telecom applications, improving customer support efficiency by 30%. Architected RAG pipelines using LangChain, Azure OpenAI, PostgreSQL, embeddings, and vector search to improve retrieval accuracy by 25% and reduce manual search effort. Engineered FastAPI-based AI microservices and REST APIs for secure, low-latency inference. Designed agentic AI workflows using LangGraph, Azure OpenAI, and tool calling to automate multi-step business processes with improved orchestration and reliability. Optimized retrieval pipelines and prompt engineering to reduce response latency by 35% while improving relevance. Built LLM evaluation frameworks, AI guardrails, and human-in-the-loop validation to reduce hallucinations. Developed data pipelines (PySpark/SQL/Pandas/Snowflake) for high-volume telecom datasets and deployed/optimized Azure workloads with Docker, Kubernetes, and CI/CD to reduce deployment time by 40%.

AI Engineer at Cognizant

Developed and deployed AI-powered applications using Python, FastAPI, and Azure OpenAI, improving business process efficiency. Built LLM-powered assistants with LangChain and prompt engineering for document understanding, question answering, and enterprise knowledge retrieval. Designed scalable REST APIs and backend services to integrate AI models securely and reliably. Developed NLP pipelines for classification, entity extraction, semantic search, and summarization, reducing manual processing by 30% and improving information discovery. Implemented ML models using Scikit-learn and XGBoost, improving prediction accuracy by 18%. Improved model reliability through preprocessing, feature engineering, and evaluation workflows. Containerized AI applications with Docker and deployed to Azure, enhancing consistency and scalability, and contributed to CI/CD automation, testing, and monitoring.