May Chitayat Email & Phone Number

End-to-end model development and pipeline integration for deep learning applications. Specific projects included:• Leading a few-shot learning project, implementing cutting-edge research to reduce data requirements and enhance model efficiency.• Developing and optimizing deep learning models for X-ray image classification.• Implementing a Transformer-based framework for semantic segmentation (Mask2Former).• Conducting a time-series prediction project, accurately forecasting load hours and shifts to reduce SLA violations and improve operational efficiency.• Leveraging tools like Ray Tuner and W&B to optimize the training and evaluation process, including hyperparameter tuning, achieving significant improvements in model performance.• Collaborating with cross-functional teams to deploy models into production environments using AWS (S3, SageMaker), Docker, and CI/CD pipelines.

Providing consultation, POC and end-to-end model pipeline integration for several client companies,all with extensive use of Python and SQL complex operations. Specific projects included:• Using classical ML models for classification and regression (Random Forest, Logistic regression, XGBoost) for large scale (100+ Million events) Real-time bidding. Working directly with Data engineering team for model integration and real-time performance.• Setting up a recommendation system for live streaming services using similarity search engine (FAISS) for near real-time recommendation at large scale.

Tel Aviv, Israel

• Responsible for MF IBM Z/OS various major tasks using Control-M, Jes2, JCL, VTAM, CICS, IPL, TSO.• Working under continuous stress with the highest standards.