Open to opportunities

Zineb h

@zinebh

Message

I'm building multimodal AI systems for clinical reporting, NLP applications, and full-stack healthcare workflows.

Algeria

Message

What I'm looking for

I'm looking to build practical AI systems, especially in healthcare, NLP, computer vision, and efficient deep learning, while continuing to grow through research and full-stack product development.

I'm developing an automated medical report generator with Ibn Al Haythem Ophthalmology Center, using 170,000+ patient records to produce structured French-language ophthalmic reports. I trained a multimodal EfficientNet-B0 model that combines fundus images and 94 clinical measurements, then integrated an LLM critique-then-revise pipeline to reduce hallucinations.

I'm also a Ph.D. student in Artificial Intelligence at ENSIA researching cost- and energy-aware compression of deep neural networks and large language models. Previously, I built a bilingual sentiment-analysis web application at Cevital SPA and delivered projects in IoT attack detection, ECG classification, RAG-based learning, forecasting, and full-stack web development.

Experience

Work history, roles, and key accomplishments

Current

Ph.D. Student, Artificial Intelligence

Current

National Higher School of Artificial Intelligence (ENSIA)

Jan 2026 - Present (7 months)

Pursuing a Ph.D. in Artificial Intelligence with a thesis on cost- and energy-aware compression of deep neural networks and large language models.

Current

State Engineering Degree & Master's, Artificial Intelligence & Data Science

Current

National Higher School of Artificial Intelligence (ENSIA)

Jan 2021 - Present (5 years 7 months)

Pursuing a State Engineering Degree and Master's in Artificial Intelligence and Data Science.

Final-Year Project – Automated Medical Report Generator

Ibn Al Haythem Ophthalmology Center

Sep 2025 - Jun 2026 (9 months)

Built an end-to-end AI system generating structured French-language ophthalmic reports from fundus images and clinical measurements, using 170,000+ real patient records. Integrated an LLM-based report generation pipeline with a Critique-then-Revise mechanism to detect and correct hallucinations.

EfficientNetB0 Multimodal Fusion LLM Full Stack Web Development Flask SQL Python

Software Engineering Intern

Cevital SPA

Sep 2024 - Oct 2024 (1 month)

Built a full-stack bilingual sentiment-analysis web application with user/admin dashboards, campaign and announcement management, role-based access control, and automated sentiment reporting.

Flask SQL Sentiment Analysis Role Based Access Control Web Development