Distributed ML Engineer: Real-Time Inference & GPU Training

IMC Trading; Hong Kong; Sur place; HKD 700 000 - 1 100 000; Plein temps

Résumé du poste

IMC Trading invites a Machine Learning Engineer to join a team building scalable ML training and inference systems that power global trading operations. You will work with researchers, hardware experts, and software engineers to push GPU acceleration and distributed computing to the limit.

You will help shape models used in live trading, optimize pipelines, and integrate cutting-edge open-source tools across our production stack. The role combines research, engineering, and production impact.

Qualifications

• 3+ years of experience in machine learning with a focus on training or inference systems.
• Hands-on experience with real-time, low-latency ML pipelines in high-performance environments is a strong plus.
• Strong engineering skills, including Python, CUDA, or C++.
• Knowledge of ML frameworks such as PyTorch, TensorFlow, or JAX.
• Proficiency in GPU programming for training and inference acceleration (CuDNN, TensorRT).
• Experience with distributed training for scaling ML workloads (Horovod, NCCL).
• Exposure to cloud platforms and orchestration tools.
• A track record of contributing to open-source projects in ML, data science, or distributed systems is a plus.

Responsabilités

• Develop large-scale distributed training pipelines to manage datasets and complex models.
• Build and optimize low-latency inference pipelines, delivering real-time predictions in production.
• Develop libraries to improve ML framework performance.
• Maximize training and inference performance using GPU hardware and acceleration libraries.
• Design scalable model frameworks for high-volume trading data with real-time, high-accuracy predictions.
• Collaborate with researchers to automate ML experiments, hyperparameter tuning, and retraining.
• Partner with HPC specialists to optimize workflows, speed up training, and reduce costs.
• Evaluate and roll out third-party tools to enhance model development, training, and inference capabilities.
• Dig into open-source ML tools to extend capabilities and improve performance.

Connaissances

Python

CUDA

C++

PyTorch

TensorFlow

JAX

CuDNN

TensorRT

Horovod

NCCL

Distributed training

Cloud platforms

Outils

Horovod

NCCL

CuDNN

TensorRT

Description du poste

IMC Trading invites a Machine Learning Engineer to join a team building scalable ML training and inference systems that power global trading operations. You will work with researchers, hardware experts, and software engineers to push GPU acceleration and distributed computing to the limit.

You will help shape models used in live trading, optimize pipelines, and integrate cutting-edge open-source tools across our production stack. The role combines research, engineering, and production impact.