Remote Distributed Training Engineer — Multi-GPU Scaling

BairesDev; Mexico; On-site; PHP 7,299,000 - 11,557,000; Full time

Benefits offered by this job

100% remote work

Excellent USD compensation

Hardware and software setup

Flexible hours

Paid parental leaves and holidays

Multicultural global team

Mentorship and growth opportunities

Job summary

BairesDev, a leading technology company, seeks a Distributed Training Engineer to scale and optimize multi-GPU, multi-node training workloads. You will work on mixed-precision strategies and coordinate across nodes to maximize throughput, collaborating with ML research teams worldwide.

The role emphasizes deep experience with PyTorch, NVIDIA tooling, and inference optimizations, with a strong focus on performance and scalability in a remote, globally distributed environment.

Qualifications

• 4+ years of experience in Machine Learning Engineering or Distributed Systems.
• Proven expertise in distributed training frameworks like PyTorch DDP, FSDP, or DeepSpeed.
• Hands-on experience with Nvidia GPU profiling tools such as Nsight.
• Strong proficiency in CUDA programming and performance optimization for multi-node clusters.
• Experience with inference optimization tools like Triton or TensorRT.
• Advanced proficiency in English.

Responsibilities

• Implement and optimize distributed training strategies using PyTorch DDP, FSDP, and DeepSpeed.
• Profile training performance and GPU utilization using Nvidia ecosystem tools like Nsight.
• Design mixed-precision and memory-optimization workflows to maximize training throughput and reduce memory footprint.
• Solve complex coordination and synchronization issues across multi-node clusters using CUDA and Triton.
• Scale massive models through parameter sharding, pipeline parallelism, and efficient data sharding techniques.
• Partner with machine learning research teams to align infrastructure capabilities with evolving model scaling requirements.

Skills

Distributed training

PyTorch DDP

FSDP

DeepSpeed

CUDA

Triton

English

Tools

Nsight

TensorRT

Job description

BairesDev, a leading technology company, seeks a Distributed Training Engineer to scale and optimize multi-GPU, multi-node training workloads. You will work on mixed-precision strategies and coordinate across nodes to maximize throughput, collaborating with ML research teams worldwide.

The role emphasizes deep experience with PyTorch, NVIDIA tooling, and inference optimizations, with a strong focus on performance and scalability in a remote, globally distributed environment.