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 developing machine learning, NLP, and generative AI solutions across banking and healthcare domains. Skilled in Python, PyTorch, AWS, and Azure Machine Learning, with hands-on expertise in RAG pipelines, model optimization, explainability, MLOps, and responsible AI.…

Riddhima Reddy Ramasahayam

AI Engineer, Data Scientist, Data Engineer, +1





AI/ML Engineer with 3+ years of experience developing machine learning, NLP, and generative AI solutions across banking and healthcare domains. Skilled in Python, PyTorch, AWS, and Azure Machine Learning, with hands-on expertise in RAG pipelines, model optimization, explainability, MLOps, and responsible AI.…

Available to hire

AI/ML Engineer with 3+ years of experience developing machine learning, NLP, and generative AI solutions across banking and healthcare domains. Skilled in Python, PyTorch, AWS, and Azure Machine Learning, with hands-on expertise in RAG pipelines, model optimization, explainability, MLOps, and responsible AI.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Beginner

Work Experience

AI/ML Engineer at Fifth Third Bank

August 1, 2025 - Present

Built an enterprise Banking Decision Copilot for lending, risk, compliance, and operations using RAG, policy retrieval, and citation-backed response generation, reducing manual policy research time by 60%. Engineered scalable ETL pipelines to ingest structured and unstructured sources (SharePoint, Confluence, SAP, Hadoop, PostgreSQL, PDFs), enabling classification, metadata extraction, deduplication, chunking, embedding generation, and vector indexing across 1.5M+ chunks. Designed a multi-agent RAG workflow with LangChain/Llama/GPT-4, FAISS, and rerankers to improve retrieval relevance, and implemented AI governance controls (confidence scoring, source validation, freshness checks, citation verification, guardrails, and human-in-the-loop) reducing unsupported recommendations by 18%. Deployed production inference services with FastAPI, Docker, Kubernetes, AWS EC2/Lambda and GitHub Actions CI/CD, optimizing latency, reliability, and inference cost (reduced costs by 40%) using caching and

ML Engineer at Genpact

January 1, 2022 - July 31, 2024

Developed an end-to-end population health analytics pipeline transforming claims, demographics, and clinical data into patient risk insights for care-management teams. Translated domain requirements into validation rules, feature engineering logic, and model specifications across Agile cycles. Built Python/SQL ETL pipelines to extract, cleanse, validate, and transform healthcare data from MySQL to support reliable features for risk modeling. Created BERT-based clinical NLP pipelines to extract semantic features from unstructured notes, improving patient risk profiling. Built predictive modeling workflows using scikit-learn and XGBoost (plus ensembles and segmentation) to identify high-risk cohorts, improving F1 score by 18%. Managed experiments with MLflow (tracking 25+ runs) and applied SHAP/LIME for explainability. Containerized models with Docker and supported AWS-based MLOps packaging, versioning, and deployment preparation, including documentation for production readiness.