Open to opportunities

Ilay Sabah

@ilaysabah

Message

Data Scientist building end-to-end healthcare AI with NLP and multimodal ML.

Israel

Message

What I'm looking for

I’m looking to build healthcare AI products—end-to-end predictive modeling, NLP and multimodal systems—turning clinical data into actionable insights. I want hands-on work with modern LLM/RAG tooling and scalable data pipelines.

I build end-to-end AI systems for healthcare, turning complex clinical data into high-impact insights and scalable ML solutions. At DataMED Lab - TAU, I led predictive analytics with a major Israeli HMO, analyzing ~1,700 patients and 65+ features to identify key risk factors for disease outcomes using HuggingFace, fine-tuning, and RAG-style approaches.

I also develop deep learning NLP and multimodal frameworks for clinical prioritization, including work analyzing data from over 7,000 patients across 7 countries. I engineer robust data pipelines (ETL, Spark), vector/data infrastructure (FAISS, ChromaDB, MongoDB), and hybrid multimodal architectures combining RAG-augmented context with tabular and graph-based relational features (GCN), while strengthening model performance and reliability through careful preprocessing and tooling.

Experience

Work history, roles, and key accomplishments

Current

Data Scientist & ML Engineer

Current

DataMED Lab - TAU

Jul 2025 - Present (1 year 1 month)

Led an end-to-end predictive analytics project with a major Israeli HMO, analyzing ~1,700 patients to identify key risk factors for disease outcomes using ML and LLM techniques. Built deep learning NLP and multimodal architectures for medical exam prioritization and clinically grounded prediction, including RAG-based context and graph-based modeling.

Freelance Data Annotator

Data Annotation Tech, Outlier, Peroptyx

Sep 2024 - Jul 2025 (10 months)

Crafted and refined high-quality data prompts and annotations across Python, Java, JavaScript, mathematics, and general knowledge topics. Reviewed and corrected peer annotations to maintain consistent quality and compliance with guidelines.

Data Annotation Python Java JavaScript Mathematics

Education

Degrees, certifications, and relevant coursework

Holon Institute of Technology

Bachelor of Science, Digital Medical Technologies

2024 -

Grade: GPA: 91

Multidisciplinary B.Sc. program at the intersection of healthcare and technology, focused on data engineering, computer science, and medical knowledge. Coursework includes machine learning & deep learning, NLP, data mining, computer vision, and big data architectures.