Lead Machine Learning Engineer, Foundation Models
Core
Lead the design, pre-training, and adaptation of massive, multi-modal foundation world models to simulate physical environments and enable spatio-temporal reasoning for Grab's marketplace.
Role type
Senior IC Lead Machine Learning Engineer (Foundation Models)
Builds
Generalized, foundational architectures for zero-shot and few-shot spatio-temporal reasoning and action-conditioned environment simulation.
Domain
Generative AI, Foundation Models, Reinforcement Learning
Deliverable
production ML models
Required skills
Large Language Models (LLMs), Vision-Language Models (VLMs), self-supervised learning architectures, distributed training pipelines, multi-billion parameter model training, petabyte-scale data pipeline architecture, Python, PyTorch or JAX
Preferred skills
State-space models (e.g., Mamba), latent variable models, diffusion models, transformer architectures
Technologies
Ray, FSDP, DeepSpeed, multi-node GPU clusters
Responsibilities
Lead end-to-end development of foundational world models and make critical architectural decisions; Design and optimize distributed training pipelines for massive multimodal datasets; Translate state-of-the-art academic papers into scalable, production-ready code; Optimize model inference for low-latency and high-throughput; Mentor team members on foundation world model development.
Seniority
Senior, hands-on IC with technical leadership
