Led GenAI-powered clinical assistant (ReX-Health) using GPT-4, Gemini, LangChain, with RAG, Pinecone, and LlamaIndex to improve clinical query accuracy by 35% and physician adoption by 30%. Engineered scalable vector search pipelines with FAISS, ChromaDB, and OpenAI embeddings enabling sub-second retrieval across 50K+ patient documents. Implemented automated CI/CD with Docker, MLflow, and GitHub Actions; containerized multi-LLM microservices via FastAPI and Kubernetes with autoscaling on AKS. Deployed end-to-end monitoring with Prometheus and Grafana; integrated Azure Cognitive Search for hybrid retrieval, boosting clinical query accuracy by 41%. Designed RLHF-style feedback loops with clinician validation and bandit algorithms to improve reliability.

Machine Learning Engineer at Hellinex Cloud

December 1, 2020 - July 1, 2023

Developed digital twin reinforcement learning models (PPO, DDPG) for irrigation optimization; built real-time soil anomaly detection with YOLOv5, OpenCV, and QGIS overlays; automated ingestion pipelines with Snowflake and Apache Spark; authored an internal whitepaper on causal inference with DoWhy, EconML, SHAP; delivered Power BI dashboards with prescriptive AI insights increasing operational efficiency by 25%.

Machine Learning Engineer at Parallax Pros

June 1, 2019 - November 1, 2020

Developed CNN-powered visual defect detection on edge devices with TensorFlow/TFLite; implemented real-time fraud detection using Isolation Forests and Autoencoders; built an LSTM-based predictive maintenance framework on Azure Databricks, forecasting equipment failures with 28% higher accuracy; automated ETL workflows with Apache Airflow and DVC; optimized vector-based similarity search across MongoDB and PostgreSQL for real-time analytics.