About
I've spent eight years turning data into decisions in energy, aviation and retail.
I am especially interested in Arabic natural language processing, and in building models that actually work in production — not just in the lab.
(This is a demo site on Ta3reefy — the person and details are fictional.)
Experience
-
2021 Present
Senior ML Engineer
Najma Airways
Built a seat-demand forecasting model that lifted revenue by 4%; leading a team of five engineers.
-
2018 2021
Data Scientist
Gulf Energy Company
Developed predictive-maintenance models that reduced unplanned equipment downtime.
-
2016 2018
Data Analyst
Al Wafra Retail
Education
-
2014 2016
MSc Data Science
Hamad Bin Khalifa University
Thesis on sentiment analysis of Gulf-dialect Arabic tweets.
-
2010 2014
BSc Computer Engineering
Qatar University
My work
Forecasting airline demand
We combined five years of bookings with seasons, holidays and major events. We built a model that forecasts demand for every flight ninety d... Read moreWhy models struggle with Arabic dialects
Most models are trained on Modern Standard Arabic, but people write online in their own dialects. The same word can be positive in one diale... Read moreSkills
- Python
- Machine learning
- Deep learning
- Arabic NLP
- Time-series forecasting
- SQL
- PyTorch
- MLOps
- Data visualisation
- Statistics