Lead Data Scientist (Fulfilment)Singapore
Life at Grab
At Grab, every Grabber is guided by The Grab Way, which spells out our mission, how we believe we can achieve it, and our operating principles - the 4Hs: Heart, Hunger, Honour and Humility. These principles guide and help us make decisions as we work to create economic empowerment for the people of Southeast Asia.
Get to know the Fulfillment
The Fulfillment tech family is one of the pillars enabling Grab to out-serve our consumers and partners in different businesses and marketplaces across Southeast Asia. We are working on high throughput, real-time distributed systems that use sophisticated machine learning techniques to solve hundreds of millions of requests per day. Our mission is to offer the best-in-class products and experiences to our driver partners as to increase adoption and engagement of our services. Improve driver partner opportunities and efficiency in order to fulfill consumer orders without fail, rain or shine. And to create efficient marketplaces by determining an optimal price that is both sustainable and loved by our partners and consumers.
Get to know the Team
Grab is Southeast Asia’s leading super-app. We provide everyday services such as deliveries, mobility, financial services, enterprise services and others to millions of users across the region. At the fulfillment machine engineering team, we are trying to solve challenging problems in the marketplace that involve dynamic pricing, supply and demand management. We are looking for senior machine learning engineers to join the team to help us make that vision a reality by developing and refining cutting-edge reinforcement learning models and simulation platforms.
Get to know the Role
This is a hands-on role involving building large-scale simulation platforms and reinforcement algorithms. You will have the opportunity to build a digital twin of Grab’s marketplace that consists of tens of thousands of consumers, drivers and merchants. Furthermore, you will have the opportunity to develop reinforcement learning, optimization and control models to solve business problems inside Grab’s marketplace and deploy them at scale.
The ideal candidate will have solid understanding of software development life-cycle and engineering practices, experience developing production ML systems, experience working on a range of regression/classification and optimization problems, experience applying reinforcement learning (or control theory), experience working with real-time streaming data.
The Day-to-Day Activities
Architect and develop our simulation platform to simulate the response from the real marketplace that involves different types of services that Grab is providing
Collaborate with product analysts, managers and business teams to define, prototype and build simulation SDKs to facilitate users to run simulations under different scenarios
Architect and develop reinforcement learning frameworks to train and run reinforcement learning algorithms at scale, provide efficient optimization solutions to challenging business problems such as pricing, demand and supply management.
Collaborate with data science and economists teams to solve difficult causality and behavior modeling problems for building more efficient simulation and reinforcement learning algorithms
Engage in service capacity and demand planning, software performance analysis, costing, tuning and optimization.
Participate in code and design reviews to maintain our high development standards.
A degree in computer science, software engineering, information technology or related fields
5+ years of experience in one or more of the following areas: general machine learning, deep learning, reinforcement learning (control)
Solid understanding of engineering practices and design patterns, experience in writing readable, maintainable and testable code
Familiarity working with VCS such as git, git-flow, understanding of full software development life-cycle
Experience turning business problems into ML/AI-projects
Experience with any ML framework, such as TensorFlow or PyTorch
Proficiency in Python
Experience in any of Scala/Java/Golang/C++
Experience with any big data framework, such as Spark, Ray familiar with the concept of processing events in real-time
The Nice to Have
A Masters or PhD in computer science, machine learning or related fields
Experience working with streaming data using Apache Flink, Apache Spark
Experience developing production quality ML Pipelines
Experience with MLFlow, TensorFlow probability, TensorFlow agents, Ray RLLib
Experience with distributed systems and cloud services (AWS, GCP, AZURE)
Experience applying reinforcement learning for solving real-world problems (robotics, finances, etc.).
Understanding of probabilistic modeling and differential programming, ability to design/build probabilistic simulators
Contributions to open source projects
We are committed to building diverse teams and creating an inclusive workplace that enables all Grabbers to perform at their best, regardless of nationality, ethnicity, religion, age, gender identity or sexual orientation and other attributes that make each Grabber unique.
Grab is an equal opportunity employer. We owe our success to the talents of our globally-diverse team and the varying perspectives they add to our thriving community.
Grab does not accept unsolicited resumes sent by recruiting agencies. Please do not forward resumes to our job postings, Grab employees or other parts of the business. Grab will not be liable to pay any fees to agencies for candidates hired as a result of unrequested resumes.