Get to Know the Team

The Fulfilment Tech Family is a foundational part of Grab, enabling seamless coordination between our diverse marketplaces across Southeast Asia. We design real-time, distributed systems and machine learning solutions to process hundreds of millions of requests per day, driving efficient supply allocation, pricing, and order matching. Our mission: Deliver best-in-class products for our driver-partners. Maximise efficiency in fulfilling consumer demand - rain or shine. Create sustainable, efficient marketplaces that balance experience and cost for all stakeholders. We're seeking a Senior Principal Machine Learning Engineer to join our Fulfilment team and take the lead in Fulfilment strategy optimization and driver behaviour modelling - a strategic area for understanding our partners and optimising our pricing, dispatch and supply management policies.

Get to Know the Role

You'll report to the Head of Engineering and work onsite at Grab's One North Singapore office. You'll focus on optimising cross-system fulfilment strategies and modelling driver responses to different platform interventions. This includes leading development in reinforcement learning (RL), behavioural prediction, and simulation-based optimization techniques, aimed at improving operations across our marketplace.

Your work will involve building interpretable, adaptable multi-agent RL systems or decision agents that consider multiple objectives while handling disruptions. You'll also develop high-fidelity models that simulate marketplace operations, aiding teams in designing smarter algorithms and driving impactful product decisions.

The Critical Tasks You Will Perform

You will:

• Apply advanced ML/ DL models to enhance performance and improve generalisation and efficiency
• Develop unified RL architectures that coordinate multiple levers (pricing, dispatching, and supply planning) with differing objectives and time scales.
• Build multi-agent or hierarchical RL frameworks to jointly optimize pricing, dispatching, and repositioning decisions, pushing the marketplace Pareto frontier.
• Research scalable representations of marketplace state that incorporate supply-demand signals, elasticity, traffic, weather, and driver intent
• Develop robust, interpretable models to capture driver decision-making under varying operational conditions.
• Design feedback loops that adapt to driver behaviour over time and across geographies.
• Collaborate with platform and experimentation teams to run real-world validations and iterate on model design.
• Build tooling and simulations to support counterfactual analysis and platform design decisions.
• Operate as a technical lead, guiding data scientists in these efforts while fostering a collaborative and high-performance environment
• Work with data engineers and backend engineers to integrate optimization models into real-time production systems.
• Support the broader roadmap of supply planning, pricing, dispatch, and marketplace experimentation across Fulfilment.

What Essential Skills You Will Need

• You have a Master's degree in Computer Science, Operations Research, Applied Mathematics, or related field with at least 10 years of relevant experience.
• You have experience with Reinforcement Learning, MDPs, stochastic control, or behavioural modelling under uncertainty.
• You have experience developing and deploying ML models that incorporate online learning, temporal decision processes, or simulation-based optimization.
• You are fluent in Python and ML frameworks (e.g., PyTorch, TensorFlow).
• You are familiar with distributed computing systems or scalable training platforms (e.g., Spark, Ray).
• You can translate high-level business problems into tractable modelling tasks.