🛠 Tech Stack

💼 About This Role

You'll design and optimize core training infrastructure for frontier AI models, bridging research and production. Your impact includes building RL training loops, distributed GPU systems, and petabyte-scale data pipelines. You'll work on open superintelligence that is accessible to all.

🎯 What You'll Do

• Design and optimize large-scale training loops and data pipelines.
• Implement state-of-the-art techniques with numerical stability.
• Build internal tooling for launching and monitoring experiments.
• Diagnose bottlenecks across the training stack (GPU, communication, dataloader).

📋 Requirements

• Strong software engineering experience with ML systems.
• Deep experience in distributed training & inference or data infrastructure.
• Ability to implement research papers into production-grade code.
• Experience with performance optimization in distributed systems.

✨ Nice to Have

• Experience with NCCL, RDMA, GPU interconnects.
• Familiarity with FSDP / ZeRO and model sharding.
• Knowledge of Triton or custom kernel development.

🎁 Benefits & Perks

• 💰 Top-tier compensation: Salary and equity globally competitive.
• 🏥 Health & wellness: Comprehensive medical, dental, vision, life, disability.
• 👶 Life & family: Paid parental leave, family planning support.
• 🌴 Paid time off and relocation support.
• 🍽️ Daily meals and team off-sites.