AI/ML Engineer with Healthcare Focus

Summary

HARSH PARIKH

AI/ML Engineer

USA +1-219-***-**** Email: ***************@*****.*** LinkedIn Results-driven AI/ML Engineer with 3 years of experience designing, building, and deploying machine learning, NLP, and deep learning models across healthcare and enterprise domains. Skilled in Python, TensorFlow, PyTorch, Scikit-learn, and AWS SageMaker, with strong expertise in data pipelines, model optimization, and MLOps automation. Adept at translating complex business needs into scalable AI solutions that improve accuracy, efficiency, and decision-making.

Professional Experience

AI/ML Developer Universal Health Services, (Internship) USA Dec 2024 – Present

• Designed and deployed deep learning models using TensorFlow and PyTorch to predict patient readmission risks and optimize care pathways, improving clinical decision accuracy by 24%.

• Developed end-to-end ETL and ML pipelines with Apache Airflow, Spark, and Snowflake to process 15 TB+ of healthcare data from EHR, claims, and IoT sources, reducing data latency by 40%.

• Built and fine-tuned BERT-based NLP models for medical note classification, entity recognition, and clinical report summarization, achieving F1-scores above 0.91 across multiple data domains.

• Deployed real-time ML inference APIs on AWS SageMaker and FastAPI for hospital dashboards, reducing processing delays to under 100 ms while scaling to 5,000+ concurrent users.

• Partnered with data engineers to establish automated MLOps pipelines using MLflow, Docker, and Kubernetes, cutting model deployment time by 45% and ensuring versioned reproducibility.

• Created Power BI dashboards visualizing KPIs such as model accuracy, data drift, and operational throughput, improving visibility for data science leadership and executives.

• Led data quality validation and bias detection frameworks ensuring compliance with HIPAA, SOC2, and internal governance standards, mitigating risks in model predictions.

• Collaborated with cross-functional healthcare stakeholders to integrate AI outputs into clinical systems (Epic, Cerner), streamlining diagnosis and decision support workflows across facilities.

• Mentored junior engineers on best practices in PyTorch, MLOps automation, and cloud-based experimentation, improving team delivery velocity by 30%

Machine Learning Engineer Hexaware Technologies India April 2021 – June 2023

• Designed AI-driven automation solutions for banking and insurance clients using Scikit-learn and XGBoost, improving document classification accuracy by 35% and reducing manual review time by 50%.

• Built predictive ML pipelines with Python and SQL to forecast customer churn and claim fraud, achieving AUC > 0.92 and driving $1.2M annual savings through proactive intervention.

• Developed computer vision systems with OpenCV and YOLOv7 for document OCR and identity verification, automating validation processes with 95% precision.

• Implemented MLOps pipelines integrating Airflow, MLflow, and Docker for reproducible model tracking, automated retraining, and deployment to Azure ML, reducing downtime by 40%.

• Created data ingestion and transformation workflows using Spark and Kafka to process structured and unstructured data from APIs and flat files, ensuring scalability and real-time analytics.

• Collaborated with the data architecture team to optimize Snowflake-based feature stores, improving model feature access latency by 30%.

• Applied hyperparameter tuning techniques (Optuna, GridSearchCV) and ensemble stacking to improve predictive accuracy across multiple ML tasks.

• Documented architecture diagrams, feature engineering workflows, and API design specifications in Confluence, improving onboarding and project traceability.

• Partnered with business stakeholders to define KPI metrics and deliver interactive Tableau dashboards tracking AI impact across risk and operations teams.

Technical Skills

• Programming: Python (Pandas, NumPy, Scikit-learn, TensorFlow, PyTorch), R, Java (basic), SQL

• Machine Learning: Regression, Classification, Clustering, Ensemble Methods, Transfer Learning

• Deep Learning: CNN, RNN, LSTM, GANs, Autoencoders, Attention Models, Transformers

• NLP & GenAI: Hugging Face Transformers, BERT, GPT-4, spaCy, NLTK, Summarization, Sentiment Analysis

• Computer Vision: OpenCV, YOLOv8, Detectron2, Segmentation Models, OCR (EasyOCR, Tesseract)

• Data Engineering: Apache Spark, Databricks, Kafka, Snowflake, Airflow, Azure Data Factory

• MLOps & DevOps: MLflow, Kubeflow, Docker, Kubernetes, CI/CD (Jenkins, GitHub Actions), DVC

• Cloud Platforms: AWS (SageMaker, EC2, Lambda, S3), Azure ML, GCP Vertex AI

• Big Data & Storage: Hadoop, Hive, Redshift, BigQuery, S3, PostgreSQL, MongoDB

• APIs & Model Deployment: FastAPI, Flask, REST APIs, ONNX, TorchServe, Triton Inference Server

• Visualization: Power BI, Tableau, Seaborn, Matplotlib, Plotly

• Collaboration Tools: Agile, JIRA, Confluence, ServiceNow, Git Certifications

• Build a blockchain and cryptocurrency

• Machine Learning A-Z & Deep Learning A-Z

Education

Master in Computer Science Purdue University, Fort Wayne, IN, USA Aug 2023 – May 2025 Bachelor of Information Technology Thadomal Shahani Engineering College, India Aug 2019 – May 2023

Contact this candidate