Senior Applied Engineer
HybridSingapore, Singapore
Job Summary
Design and build physical foundation models by efficiently scaling training to large datasets on multi-GPU cloud compute. Transform research prototypes into robust, optimized implementations while challenging architecture decisions that hinder scalability. Work with scientists, engineers, and customers to deliver AI models addressing real-world physics and engineering problems across aerospace, defense, and automotive sectors. Identify optimal libraries, frameworks, and tools for the research team, then translate results into reusable libraries, tooling, and products. Own research work-streams at various levels, fostering curiosity and initiative among colleagues and mentees. Discuss work results and implications with stakeholders to demonstrate how solutions address industry challenges.
Required Qualifications
- MSc or PhD in computer science, machine learning, applied statistics, mathematics, physics, engineering, software engineering, or a related field
- record of experience in scientific computing
- record of experience in high-performance computing (CPU / GPU clusters)
- record of experience in parallelised / distributed training for large / foundation models
- exposure to scaling and optimising ML models, training and serving foundation models at scale
- exposure to distributed computing frameworks (e.g., Spark, Dask) and high-performance computing frameworks (MPI, OpenMP, CUDA, Triton)
- exposure to cloud computing (on hyper-scaler platforms, e.g., AWS, Azure, GCP)
- exposure to building machine learning models and pipelines in Python, using common libraries and frameworks (e.g., NumPy, SciPy, Pandas, PyTorch, JAX), especially including deep learning applications
- exposure to C/C++ for computer vision, geometry processing, or scientific computing
- exposure to software engineering concepts and best practices (e.g., versioning, testing, CI/CD, API design, MLOps)
- exposure to container-ization and orchestration (Docker, Kubernetes, Slurm)
- exposure to writing pipelines and experiment environments, including running experiments in pipelines in a systematic way
- Ability to work autonomously and scope and effectively deliver projects across a variety of domains
- Strong problem-solving skills and the ability to analyse issues, identify causes, and recommend solutions quickly
- Excellent collaboration and communication skills
Desired Qualifications
- exposure to federated learning
- Ideally, >2 years of experience in a data-driven, professional setting
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.