Ingénieur.e sénior en apprentissage automatique distribué

Résumé du poste

The role involves collaborating with machine learning researchers to accelerate model training and inference, focusing on large-scale models in distributed computing environments. Key tasks include investigating performance bottlenecks, profiling research code, debugging issues, optimizing resource utilization, and developing software tools to simplify distributed computing workflows.

Détails du poste

The role involves collaborating with machine learning researchers to accelerate model training and inference, focusing on large-scale models in distributed computing environments. Key tasks include investigating performance bottlenecks, profiling research code, debugging issues, optimizing resource utilization, and developing software tools to simplify distributed computing workflows. A degree in a relevant computer science field is required, with a postgraduate degree in ML or distributed ML systems being preferred but not mandatory with exceptional experience. Candidates must have over 3 years of experience designing distributed ML training frameworks, proficiency with tools like Megatron/DeepSpeed/FSDP, and familiarity with cloud platforms, workload managers, and GPU profiling tools.

Ce que vous ferez

The role involves collaborating with machine learning researchers to accelerate model training and inference, focusing on large-scale models in distributed computing environments. Key tasks include investigating performance bottlenecks, profiling research code, debugging issues, optimizing resource utilization, and developing software tools to simplify distributed computing workflows.

Exigences

A degree in a relevant computer science field is required, with a postgraduate degree in ML or distributed ML systems being preferred but not mandatory with exceptional experience. Candidates must have over 3 years of experience designing distributed ML training frameworks, proficiency with tools like Megatron/DeepSpeed/FSDP, and familiarity with cloud platforms, workload managers, and GPU profiling tools.

Avantages

• Health Insurance
• Mental Health And Wellness Management Account
• Vacation Days
• Retirement Savings Contribution

Compétences indiquées

• Software · Souhaitée
• Sens de l’organisation · Souhaitée
• Training · Souhaitée
• Souci du détail · Souhaitée
• Health · Souhaitée
• management · Souhaitée
• Apprentissage automatique · Souhaitée

Autres compétences pertinentes

Relevées dans la description du poste. Confirmez les exigences importantes ci-dessus.

• Distributed ML Training
• Inference Optimization
• Performance Bottleneck Investigation
• Code Profiling
• Resource Optimization
• Software Tool Development
• Distributed Computing Orchestration
• ML Model Development Workflows
• Cloud Platforms
• Workload Managers
• GPU Profiling Tools
• Containerization Tools
• Vector Databases
• Deep Learning Research Contributions

Domaines d’emploi

• Science & Research
• Engineering
• Software
• Data & Analytics
• Technology

Autres postes chez LoiZéro