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

AI/ML Engineer with 3+ years of experience delivering end-to-end model development and migrating legacy ML systems to cloud-native architectures (AWS/Azure) across financial services and healthcare technology. Specialized in productionizing fraud detection and real-time inference at scale, and building generative AI/RAG solutions from prototype to deployment. Led high-impact systems including fraud detection over 10M+ daily transactions and clinical NLP for 50K+ medical notes, achieving measurable business improvements such as 22% reduction in NPAs and 65% reduction in manual review time. Strong in MLOps, LLM orchestration (RAG via LangChain/LlamaIndex), monitoring, and secure, compliant deployments.…AI/ML Engineer with 3+ years of experience delivering end-to-end model development and migrating legacy ML systems to cloud-native architectures (AWS/Azure) across financial services and healthcare technology. Specialized in productionizing fraud detection and real-time inference at scale, and building generative AI/RAG solutions from prototype to deployment. Led high-impact systems including fraud detection over 10M+ daily transactions and clinical NLP for 50K+ medical notes, achieving measurable business improvements such as 22% reduction in NPAs and 65% reduction in manual review time. Strong in MLOps, LLM orchestration (RAG via LangChain/LlamaIndex), monitoring, and secure, compliant deployments.

Nikitha Gummadi

AI Engineer, Data Scientist, Software Engineer, +2





AI/ML Engineer with 3+ years of experience delivering end-to-end model development and migrating legacy ML systems to cloud-native architectures (AWS/Azure) across financial services and healthcare technology. Specialized in productionizing fraud detection and real-time inference at scale, and building generative AI/RAG solutions from prototype to deployment. Led high-impact systems including fraud detection over 10M+ daily transactions and clinical NLP for 50K+ medical notes, achieving measurable business improvements such as 22% reduction in NPAs and 65% reduction in manual review time. Strong in MLOps, LLM orchestration (RAG via LangChain/LlamaIndex), monitoring, and secure, compliant deployments.…AI/ML Engineer with 3+ years of experience delivering end-to-end model development and migrating legacy ML systems to cloud-native architectures (AWS/Azure) across financial services and healthcare technology. Specialized in productionizing fraud detection and real-time inference at scale, and building generative AI/RAG solutions from prototype to deployment. Led high-impact systems including fraud detection over 10M+ daily transactions and clinical NLP for 50K+ medical notes, achieving measurable business improvements such as 22% reduction in NPAs and 65% reduction in manual review time. Strong in MLOps, LLM orchestration (RAG via LangChain/LlamaIndex), monitoring, and secure, compliant deployments.

Available to hire

AI/ML Engineer with 3+ years of experience delivering end-to-end model development and migrating legacy ML systems to cloud-native architectures (AWS/Azure) across financial services and healthcare technology. Specialized in productionizing fraud detection and real-time inference at scale, and building generative AI/RAG solutions from prototype to deployment.

Led high-impact systems including fraud detection over 10M+ daily transactions and clinical NLP for 50K+ medical notes, achieving measurable business improvements such as 22% reduction in NPAs and 65% reduction in manual review time. Strong in MLOps, LLM orchestration (RAG via LangChain/LlamaIndex), monitoring, and secure, compliant deployments.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

Work Experience

AI/ML Engineer at Stripe

March 1, 2024 - Present

Led end-to-end fraud detection pipeline on AWS SageMaker for payment transaction monitoring (10M+ daily transactions), improving precision by 21% and reducing false positives by 35%. Built and productionized document intelligence using fine-tuned LayoutLMv3 to extract structured data from 50K+ financial documents, reducing manual review time by 65% and saving 400+ analyst hours monthly. Architected a RAG application with LangChain, GPT-4, and Pinecone for internal policy/API Q&A achieving 91% response accuracy and reducing resolution time by 55%. Implemented real-time merchant risk stratification using gradient-boosted models on AWS Kinesis, reducing chargeback losses by 18%. Established MLOps workflows with MLflow and Docker for retraining/versioning and drift monitoring; integrated inference into FastAPI/Kubernetes microservices with sub-100ms latency and reliable 10M+ daily requests. Ensured PCI-DSS/SOC2 compliance with encryption, anonymization, and RBAC.

Machine Learning Engineer at Axis Bank

June 1, 2022 - July 31, 2023

Developed and deployed credit risk scoring models on 2M+ customer records using XGBoost and Random Forest, improving AUC from 0.74 to 0.89 and reducing NPAs by 22%. Built BERT-based NLP pipelines for sentiment analysis and complaint classification over 10,000+ daily interactions, reducing resolution time by 40%. Designed real-time fraud detection on Kafka streams with Isolation Forest and autoencoders (96% precision, sub-200ms latency). Built churn prediction models that supported retention campaigns reducing quarterly churn by 15% and retaining $3M annual revenue. Migrated on-prem ML pipelines to Azure Databricks/Azure ML Studio, reducing release cycle time from 4 weeks to 6 days. Implemented explainability with SHAP/LIME for regulatory-compliant decisions, and deployed scalable REST APIs on AKS with 99.9% uptime. Conducted A/B testing across 12+ model variants and integrated feature store workflows using Feast.