🛠 Tech Stack

💼 About This Role

You'll architect neural simulators that understand the grammar of the physical world, designing and training multi-billion parameter models on massive GPU clusters. You'll own the end-to-end training loop, diagnosing failure modes and improving data mixtures to drive measurable gains.

🎯 What You'll Do

• Architect neural simulators for long-form world simulations
• Scale training of multi-billion parameter models on huge clusters
• Own end-to-end training experiments: design, debug, iterate
• Improve data mixtures and tighten evaluation loops

📋 Requirements

• Very strong coding skills in Python, C++, or Rust
• Deep experience shipping or researching high-fidelity video models
• Ability to design from scratch and reason about scaling laws
• Comfortable managing large-scale experiments on massive GPU clusters

🎁 Benefits & Perks

• 🍕 Elite small team with radical ownership
• 🚀 Massive scope and direct responsibility
• ⚡ Fast-paced culture with rapid iteration