Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/ML Engineer with 4+ years of experience delivering end-to-end machine learning and GenAI solutions across healthcare and finance, with a strong focus on production-grade model development and deployment. Skilled in building ML systems with Python and popular frameworks (TensorFlow, PyTorch, scikit-learn), as well as GenAI/LLM solutions including RAG pipelines, prompt engineering, and agentic workflows. Experienced with MLOps and cloud-native deployments on AWS/Azure using Docker, Kubernetes, MLflow, and CI/CD.…AI/ML Engineer with 4+ years of experience delivering end-to-end machine learning and GenAI solutions across healthcare and finance, with a strong focus on production-grade model development and deployment. Skilled in building ML systems with Python and popular frameworks (TensorFlow, PyTorch, scikit-learn), as well as GenAI/LLM solutions including RAG pipelines, prompt engineering, and agentic workflows. Experienced with MLOps and cloud-native deployments on AWS/Azure using Docker, Kubernetes, MLflow, and CI/CD.

Nandini Kosgi

AI Engineer, Data Scientist, Data Engineer, +4





AI/ML Engineer with 4+ years of experience delivering end-to-end machine learning and GenAI solutions across healthcare and finance, with a strong focus on production-grade model development and deployment. Skilled in building ML systems with Python and popular frameworks (TensorFlow, PyTorch, scikit-learn), as well as GenAI/LLM solutions including RAG pipelines, prompt engineering, and agentic workflows. Experienced with MLOps and cloud-native deployments on AWS/Azure using Docker, Kubernetes, MLflow, and CI/CD.…AI/ML Engineer with 4+ years of experience delivering end-to-end machine learning and GenAI solutions across healthcare and finance, with a strong focus on production-grade model development and deployment. Skilled in building ML systems with Python and popular frameworks (TensorFlow, PyTorch, scikit-learn), as well as GenAI/LLM solutions including RAG pipelines, prompt engineering, and agentic workflows. Experienced with MLOps and cloud-native deployments on AWS/Azure using Docker, Kubernetes, MLflow, and CI/CD.

Available to hire

AI/ML Engineer with 4+ years of experience delivering end-to-end machine learning and GenAI solutions across healthcare and finance, with a strong focus on production-grade model development and deployment.
Skilled in building ML systems with Python and popular frameworks (TensorFlow, PyTorch, scikit-learn), as well as GenAI/LLM solutions including RAG pipelines, prompt engineering, and agentic workflows. Experienced with MLOps and cloud-native deployments on AWS/Azure using Docker, Kubernetes, MLflow, and CI/CD.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Beginner

Beginner

Work Experience

AI/ML Engineer at Oscar Health

August 1, 2024 - Present

Built AI-powered prior authorization and utilization management workflows processing 12M+ healthcare claims annually using Python, scikit-learn, and PyTorch. Developed predictive and deep learning models for authorization outcomes and clinical documentation analysis, achieving 91% accuracy and improving medical-necessity determinations by 29%. Implemented LangChain-based prompt orchestration for clinical reasoning workflows to improve LLM recommendation relevance. Created conversational AI pipelines for real-time provider interactions, reducing response times and improving satisfaction. Implemented dashboards to visualize utilization trends across millions of records. Used MLflow for model tracking and automated CI/CD for validation and deployment, reducing release cycles by 63%. Trained and deployed models on AWS SageMaker for large-scale clinical datasets, and leveraged a vector database for semantic retrieval of clinical guidelines and historical cases to improve contextual recomme

AI/ML Engineer at Deloitte

November 1, 2020 - July 31, 2023

Worked on an Enterprise Anti-Money Laundering (AML) and Financial Crime Detection platform. Used SQL to analyze/transform 120M+ transaction records to improve suspicious activity identification and support regulatory reporting. Developed deep learning fraud detection models with TensorFlow, achieving 93% classification accuracy and reducing enterprise-wide false positives. Implemented a RAG pipeline integrating regulatory guidance with historical investigation records to improve analyst decision support. Applied CNN and document classification models to analyze transaction patterns and automate compliance document categorization, increasing high-risk activity detection while reducing manual review. Engineered features at scale with Pandas/NumPy across 85M records. Containerized services with Docker and built secure REST APIs for real-time fraud scoring and alert generation with sub-second latency. Supported training/inference infrastructure with Azure ML, improving scalability by 49%