I am a Generative AI Engineer and Data Scientist with 10+ years of experience delivering enterprise-grade AI/ML solutions. I specialize in LLMs, RAG systems, agentic AI frameworks, and scalable MLOps on Azure and AWS, driving decision-making improvements and multimillion-dollar ROI. I lead cross-functional teams to deliver audit-ready AI platforms, mentor engineers, and align AI initiatives with compliance and business objectives across finance, healthcare, and telecom domains.

Nupur Thakur

I am a Generative AI Engineer and Data Scientist with 10+ years of experience delivering enterprise-grade AI/ML solutions. I specialize in LLMs, RAG systems, agentic AI frameworks, and scalable MLOps on Azure and AWS, driving decision-making improvements and multimillion-dollar ROI. I lead cross-functional teams to deliver audit-ready AI platforms, mentor engineers, and align AI initiatives with compliance and business objectives across finance, healthcare, and telecom domains.

Available to hire

I am a Generative AI Engineer and Data Scientist with 10+ years of experience delivering enterprise-grade AI/ML solutions. I specialize in LLMs, RAG systems, agentic AI frameworks, and scalable MLOps on Azure and AWS, driving decision-making improvements and multimillion-dollar ROI.

I lead cross-functional teams to deliver audit-ready AI platforms, mentor engineers, and align AI initiatives with compliance and business objectives across finance, healthcare, and telecom domains.

See more

Experience Level

Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
See more

Language

English
Fluent

Work Experience

Gen AI Engineer at Broad Ridge Financials
July 1, 2024 - Present
Led a cross-functional team to deploy a real-time hybrid recommendation engine on Azure Databricks & Data Factory, delivering a 40% lift in transaction value and a 50% increase in offer utilization. Architected an agentic AI framework using LangChain & LangGraph with transformer models (BERT, T5) on AKS to automate root-cause analysis and anomaly detection, cutting resolution time by 40%. Implemented Retrieval-Augmented Generation (RAG) with Azure Cognitive Search (vector search) and FAISS, improving retrieval speed and accuracy by 25% and enabling real-time financial data insights. Established scalable ingestion streams using Apache Kafka on Azure HDInsight and designed fraud-detection pipelines with anomaly-detection models for real-time alerts.
Gen AI Engineer / Data Scientist at Charter Communications
May 1, 2022 - June 30, 2024
Led the NexGen MarTech team, mentoring junior data scientists and coordinating across product, design, and compliance to deliver AI-driven campaigns for major clients, generating $5M+ in incremental revenue. Implemented conversational chatbots on AKS using CrewAI orchestration to manage bookings and inquiries, achieving a 3x improvement in bot response times. Fine-tuned BioBERT/ClinicalBERT for regulatory PDFs and literature assistants, reducing medical writer hours by 35%. Developed MLflow pipelines for automated training, deployment, and monitoring of generative models, cutting model iteration cycles by 40%. Integrated Azure Cognitive AI vector embeddings for faster literature retrieval and built LTV/retention models and sensor-data–driven predictions for clinical applications.
AI/ML Engineer/ Data Scientist at BJC Healthcare
August 1, 2020 - April 30, 2022
Led cross-functional AI/ML projects delivering fraud detection and NLP automation platforms to improve healthcare operations. Built real-time fraud-detection models on streaming data using XGBoost, Random Forests, Gradient Boosting, and Logistic Regression, preventing $1.3M in losses. Implemented scalable ETL pipelines from SQL Server to Hadoop, and used Azure Databricks/Spark for distributed feature engineering and model training on multi-terabyte datasets. Developed SpaCy/NLTK NLP pipelines with LLM-based summarization to improve accuracy by 27%. Automated CI/CD with Jenkins, Docker, and AKS; applied SMOTE and advanced resampling to handle class imbalance; migrated on-premises systems to Azure and contributed transformer-model prototypes for healthcare use cases.
Python Developer/ Data Scientist at Vernicks Technologies
May 1, 2015 - July 31, 2019
Developed Random Forest–based anomaly-detection models with SMOTE for credit-card fraud forecasting, delivering $1.3M in cost savings. Built a real-time fraud-alert system using Azure Stream Analytics and custom ML rules, significantly reducing false positives. Architected end-to-end ETL pipelines, engineered features, and implemented NLP sentiment-analysis pipelines with NLTK; created churn-predictive models and designed dashboards for revenue optimization. Automated model training and deployment with Azure ML AutoML and transitioned on-prem workloads to cloud-based Snowflake data warehousing.

Education

Master of Science in Electrical Engineering (Systems & Security) at University of South Florida, Tampa, FL
January 1, 2019 - January 1, 2021
Bachelor of Engineering in Electronics & Telecommunication at Rungta College of Engineering & Technology, CSVTU, India
January 1, 2011 - January 1, 2015

Qualifications

Add your qualifications or awards here.

Industry Experience

Financial Services, Healthcare, Telecommunications, Software & Internet, Professional Services