Descripción

ML Engineer with 3+ years building production AI systems across NLP, speech recognition, computer vision, document AI, recommendation systems, and quantum ML. I own projects end-to-end — from research and experimentation to deployment — and have delivered real solutions for clients in healthcare, e-commerce, and Industry 4.0.

Idiomas

Inglés
Bilingüe o nativo
Francés
Bilingüe o nativo
Árabe
Bilingüe o nativo

Preferencias de lugar de trabajo

Acepta trabajo presencial

Barcelona (hasta 50 km)

MultiverseComputing
ML Engineer
enero de 2026 - Hoy (7 meses)
Barcelona, España
Built end-to-end, scalable AI applications spanning data ingestion, model development, deployment, and monitoring.
Designed and shipped production AI systems with modular architecture, cloud infrastructure, and automated scaling to handle growing demand.
Machine learning FastAPI Pytorch Apache Kafka celery
Infor
DATA SCIENTIST
GRAN DISTRIBUCIÓN
julio de 2025 - enero de 2026 (6 meses)
Barcelona, España
• - Spearheaded the design and deployment of reusable ML models for demand forecasting, pricing optimization, and product recommendation across multiple client sectors, ensuring scalable solutions aligned with business goals.
• - Deployed an end-to-end recommendation engine for a client tenant, utilizing cross-sell and up-sell logic to drive a 9% increase in average order value.
ELYADATA
SOFTWARE DATA ENGINEER
octubre de 2023 - julio de 2025 (1 año y 9 meses)
Tunis, Tunisia
• - Built an AI-powered meeting assistant that uses advanced speech-to-text models to generate structured meeting summaries and insights.
• - Optimized and enhanced an ASR microservice to support multilingual and dialectal speech recognition, integrating voice activity detection and audio fragmentation to improve decoding efficiency.
• - Developed a real-time live captioning system for live streams on Youtube, integrating in-house ASR models to provide accurate transcriptions of live content.
• - Developed an OCR system specialized for Arabic language recognition, enhancing text extraction accuracy, and benchmarked our model against state-of-the-art models to evaluate and improve performance.