Senior Data Scientist

Company Description

Since launching in Kuwait in 2004, talabat has become the region’s leading on-demand delivery app, serving millions of customers across eight countries. Our Quick Commerce (QC) Hub powers the grocery and convenience delivery experience — getting everyday essentials to customers’ doors in minutes, not hours.

Behind every delivery is data. Our QC data team works at a scale most data scientists only read about: millions of daily transactions, thousands of partners, and decisions that directly shape how people across the Middle East get their groceries delivered. We’re building an analytics-first culture where every product and business decision is grounded in evidence.

Your mission: ship the AI and ML systems that power talabat's intelligence. You're AI-native by default. You use Claude, Cursor, and equivalent tools as core leverage in how you architect and deploy. This is not a research role; we measure success in business outcomes and production reliability.

You own a domain end to end — from problem framing through data pipelines, model training, deployment, and monitoring in production. You work closely with product and business teams, and you mentor other data scientists in their growth.

Responsibilities:

• Frame ambiguous business problems as well-defined ML problems with clear success criteria.
• Design, build, and ship end-to-end ML and generative AI systems in production. You handle data pipelines, feature engineering, model training, serving, and monitoring without handoffs.
• Architect robust ML-based systems. Write clean, scalable production code. Train, deploy, and maintain reliable models.
• Leverage LLMs and generative AI for data enrichment, content understanding, and automated decision-making within production systems.
• Design and analyse experiments rigorously (A/B and multivariate tests) to measure model and product impact.
• Develop deep familiarity with source data and its generating systems through collaboration with engineering teams and systematic profiling.
• Provide high-quality insights and data-driven recommendations to drive strategic organisational choices.
• Mentor other data scientists. Elevate engineering and ML best practices — improving our ways of working, tooling, MLOps, and internal training.

What we're looking for:

• 3+ years shipping production ML systems (generative AI, recommendation engines, or ML infrastructure). Demonstrated velocity and ownership.
• Strong software engineering fundamentals: clean code, solid grasp of data structures, and system design (both general and ML-specific).
• Hands-on with the ML lifecycle: problem formulation, data auditing, modelling, deployment, interpretation, presentation.
• Proficiency in Python and SQL. Comfort with reproducible analysis and experimentation (A/B and multivariate analysis). Solid statistical thinking.
• Experience building data and training pipelines (Airflow or equivalent). Strong feature engineering and data modelling skills.
• Production ML systems experience: building, deploying, serving, monitoring. You understand MLOps practices.
• Familiarity with product data (events, impressions, conversions, engagement, retention). You know how to measure product health.
• Experience with LLMs and NLP-based solutions for data enrichment and automation is a plus.
• BigQuery or equivalent cloud data warehouse experience is a plus. GCP familiarity is a plus but not required.