🛠 Tech Stack

💼 About This Role

You'll design and build large-scale GPU infrastructure for high-throughput model inference and mid-training workloads. Your work will power synthetic data generation and reinforcement learning pipelines at scale. Join a small talent-dense team building open foundation models from the ground up.

🎯 What You'll Do

• Design and build large-scale GPU infrastructure for inference and training
• Develop systems for synthetic data generation and RL pipelines
• Optimize throughput, latency, and GPU utilization for LLM inference
• Build infrastructure for distributed RL workloads and model evaluation

📋 Requirements

• Experience deploying and operating large-scale GPU systems for inference
• Several years of hands-on production infrastructure experience
• Strong understanding of GPU performance characteristics and optimization
• Experience with modern inference frameworks like SGLang or Megatron

✨ Nice to Have

• Familiarity with distributed RL infrastructure or rollout generation systems
• Experience with GPU kernels or low-level performance optimization
• Experience debugging performance issues across GPU, networking, and distributed layers

🎁 Benefits & Perks

• 💰 Top-tier compensation including salary and equity
• 🏥 Comprehensive health insurance (medical, dental, vision)
• 👶 Fully paid parental leave for all new parents
• 🍽️ Daily lunch and dinner provided
• 🌍 Relocation support and paid time off