🛠 Tech Stack

💼 About This Role

You'll design and optimize large-scale AI pre-training pipelines across distributed GPU infrastructures. You'll conduct experiments to improve model quality and efficiency, tackling computational bottlenecks in advanced language and multi-modal models. This role offers access to high-performance GPU clusters and a global, research-driven environment.

🎯 What You'll Do

• Design and optimize large-scale AI model pre-training pipelines.
• Research and prototype advanced architectures for LLMs and multi-modal AI.
• Identify and resolve bottlenecks in model performance and scalability.
• Enhance distributed training systems for reliability and efficiency.

📋 Requirements

• Large language models and transformer architectures experience.
• Distributed machine learning and GPU-based training expertise.
• Proficiency in PyTorch, TensorFlow, JAX, or similar.
• Deep learning optimization and model scaling skills.

✨ Nice to Have

• Advanced degree in CS, AI, or Machine Learning.
• Research experience with novel architectures.
• Familiarity with parallel training strategies.

🎁 Benefits & Perks

• 🏠 Fully remote work from anywhere.
• 🔬 Cutting-edge AI research on large-scale pre-training.
• 🖥️ High-performance GPU clusters access.
• 🌍 Collaborative global culture with world-class talent.
• 🚀 Fast-paced innovation-driven environment.

📨 Hiring Process

Estimated timeline: 2-4 weeks · AI estimate

1. 1Initial Screening· 30 min
2. 2Technical Interview· 60 min
3. 3Research Presentation· 45 min