Senior Data Scientist — Fraud & Trust AI

Grab; Petaling Jaya; On-site; MYR 120,000 - 180,000; Full time

Benefits offered by this job

Term Life Insurance

Medical Insurance

GrabFlex benefits

Parental leave

Birthday leave

LASA volunteering leave

Grabber Assistance Programme

Job summary

Grab is seeking a data science professional to fight fraud by analyzing transactional data and deploying scalable ML models. You will work as an Individual Contributor in a full-time, on-site role at the Petaling Jaya office, collaborating with product and software teams to push ML and LLM capabilities at scale.

You will design, train, and fine-tune model architectures using tools such as Graph Neural Networks, Transformers, and open-source LLMs, while improving data pipelines and automation.

Qualifications

• Degree in a quantitative field (e.g., CS, physics, maths).
• 3+ years of Python/SQL and data processing experience.
• Hands-on ML model development with common libraries.
• Experience with DL architectures and ML techniques.

Responsibilities

• Analyze transactional data to detect fraud with ML.
• Develop and deploy models, ensuring seamless integration.
• Collaborate with cross-functional teams across the lifecycle.
• Research advancements in algorithms and AI for fraud defence.
• Build and maintain high-fidelity training datasets and synthetic data.

Skills

Python

SQL

Spark

TensorFlow

PyTorch

XGBoost

LightGBM

Scikit-learn

Transformers

RAG

LangGraph

LangChain

Claude subagents

Education

Bachelor's degree in Computer Science/Physics/Statistics/Mathematics/Engineering/Economics

Tools

Graph Neural Networks

Transformers

GPT/LLM frameworks

Open-source LLMs (Qwen)

Job description

Grab is seeking a data science professional to fight fraud by analyzing transactional data and deploying scalable ML models. You will work as an Individual Contributor in a full-time, on-site role at the Petaling Jaya office, collaborating with product and software teams to push ML and LLM capabilities at scale.

You will design, train, and fine-tune model architectures using tools such as Graph Neural Networks, Transformers, and open-source LLMs, while improving data pipelines and automation.