Ali Abraham

Developed and deployed scalable machine learning pipelines for healthcare analytics, designing anomaly detection models to identify inconsistencies in patient data. Collaborated with DevOps to integrate model tracking and retraining pipelines using MLflow and Kubernetes, leading a team of four engineers and data scientists.

Architected a machine learning framework for real-time fraud detection in financial transactions and built a custom experimentation and evaluation platform. Implemented natural language processing models using BERT for automated support ticket classification, providing guidance on fairness, explainability, and transparency of ML systems using SHAP and LIME.

Contributed to the Adobe Sensei team to improve deep learning models used in Photoshop and Adobe Stock, enhancing smart auto-cropping and object detection features using YOLO and transfer learning. Delivered production-grade models under tight remote collaboration conditions, improving image classification accuracy by optimizing augmentation techniques.

Assisted in building machine learning models for cognitive pattern analysis in digital product testing, developing and maintaining data preprocessing pipelines using Python libraries. Supported senior engineers in training early-stage neural network models, contributing to product design improvements through insights extracted from behavior data.

Completed a Master of Science in Computer Science. Focused on advanced topics in the field, preparing for a career in machine learning and AI.