Role Summary
Own the end-to-end AI stack for four key initiatives—fraud detection, mobile-app personalization, low-code automations (Google AppSheet), and a WhatsApp chatbot. You’ll turn
raw data into production-grade models, liaise with domain experts (fraud, product, customer
support) and coordinate with third-party vendors to hit rapid short-term milestones while steering the long-term AI roadmap.
Responsibilities (Functionalities And Activities)
-
Data Engineering & Quality Develop and maintain ELT/ETL pipelines and SQL models, ensuring accurate data lineage and validation.
-
Labeling & Feature Management Design labeling strategies and manage a structured feature catalog to support model development.
-
Modeling & Experimentation Train and evaluate machine learning models, conduct A/B testing, and assess build-versus-buy decisions.
-
Deployment & Integration Deploy models as services or APIs and integrate them into applications, low-code workflows, and chat systems.
-
Monitoring & Lifecycle Management Monitor model performance (drift, latency, cost), schedule retraining, and manage rollbacks when necessary.
Qualifications
-
1-3 yrs building & deploying ML models in Python (sklearn, TensorFlow or PyTorch)
-
Solid SQL & data-pipelines (ETL, Airflow or similar)
-
Familiar with cloud services (GCP preferred; AWS/Azure fine)
-
Basic REST API design & consumption.
-
Hands-on with at least one of: fraud analytics, recommender systems, conversational AI
-
Experience with low-/no-code automation (Google AppSheet, Power Automate, Zapier, etc.)
-
Strong stakeholder communication.
-
Able to balance quick wins with scalable architectures.
- B.Sc. or higher in Data Science, CS, Engineering, Statistics, or related.
Preferred
-
Exposure to WhatsApp Business Cloud API, Twilio, or similar messaging platforms.
-
Real-time data streaming (Kafka, Pub/Sub, Kinesis)
-
MLOps tooling (Docker, Kubernetes, Kubeflow, MLflow)
-
Experience with feature stores & vector databases for personalization.
-
Arabic & English fluency.