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

I'm Harshini, an AI/ML Engineer with 5+ years of experience designing, developing, and deploying machine learning systems across healthcare and financial services. I enjoy turning complex data into actionable insights and building scalable solutions that make real business impact. I thrive in cross-functional teams, emphasize secure and compliant AI deployments, and relish opportunities to innovate with GenAI and retrieval-augmented generation to streamline workflows and improve patient outcomes.…I'm Harshini, an AI/ML Engineer with 5+ years of experience designing, developing, and deploying machine learning systems across healthcare and financial services. I enjoy turning complex data into actionable insights and building scalable solutions that make real business impact. I thrive in cross-functional teams, emphasize secure and compliant AI deployments, and relish opportunities to innovate with GenAI and retrieval-augmented generation to streamline workflows and improve patient outcomes.

Harshini Sai Sangadi

AI Engineer, AI Developer, Data Scientist, +3





I'm Harshini, an AI/ML Engineer with 5+ years of experience designing, developing, and deploying machine learning systems across healthcare and financial services. I enjoy turning complex data into actionable insights and building scalable solutions that make real business impact. I thrive in cross-functional teams, emphasize secure and compliant AI deployments, and relish opportunities to innovate with GenAI and retrieval-augmented generation to streamline workflows and improve patient outcomes.…I'm Harshini, an AI/ML Engineer with 5+ years of experience designing, developing, and deploying machine learning systems across healthcare and financial services. I enjoy turning complex data into actionable insights and building scalable solutions that make real business impact. I thrive in cross-functional teams, emphasize secure and compliant AI deployments, and relish opportunities to innovate with GenAI and retrieval-augmented generation to streamline workflows and improve patient outcomes.

Available to hire

I’m Harshini, an AI/ML Engineer with 5+ years of experience designing, developing, and deploying machine learning systems across healthcare and financial services. I enjoy turning complex data into actionable insights and building scalable solutions that make real business impact.

I thrive in cross-functional teams, emphasize secure and compliant AI deployments, and relish opportunities to innovate with GenAI and retrieval-augmented generation to streamline workflows and improve patient outcomes.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

Afar

Advanced

English

Fluent

Work Experience

AI/ML Engineer at Optum

May 1, 2024 - Present

Built an end-to-end machine learning pipeline to predict chronic disease risk using large-scale EHR and demographic data, improving early-risk identification by 18-22% over baseline models. Ingested 2M+ patient records from data lakes using SQL and Python; engineered clinical risk features and trained Random Forest and TensorFlow-based models achieving 0.82-0.86 AUC across cohorts. Designed a hybrid optimization framework combining Genetic Algorithms for population-level care prioritization and Reinforcement Learning for adaptive care plan updates, increasing intervention effectiveness by 15% in pilots. Integrated real-time patient monitoring signals with historical records to dynamically update risk scores, reducing delayed interventions by 20%. Deployed production pipelines on AWS SageMaker with SageMaker Pipelines and real-time Endpoints; dashboards with Power BI for risk stratification and treatment effectiveness.

AI/ML Engineer at Capgemini

June 1, 2020 - July 1, 2023

Developed scalable ML solutions to predict borrower default risk and segment customers for automated credit decisioning. Built ETL pipelines to ingest, cleanse, and transform financial datasets; performed feature engineering to improve signal strength. Trained and deployed supervised models using Scikit-learn, XGBoost, and LightGBM achieving 0.78-0.83 AUC and improved risk tier classification. Implemented workflow automation with Apache Airflow and model monitoring/drift detection; integrated models into loan origination systems to accelerate decision cycles by 25%. Built Tableau dashboards for risk management and decision traceability.