🛠 Tech Stack

💼 About This Role

You'll help build end-to-end 3D native machine learning systems at the market leader in 3D generative AI. Your work will directly impact foundational pretraining, finetuning, and inference for 3D models used by enterprises like Meta and millions of end users.

🎯 What You'll Do

• Co-design novel 3D model architectures with researchers
• Optimize distributed model training across hundreds to thousands of GPUs
• Build efficient multi-stage inference pipelines with compilation and quantization
• Implement and maintain 3D specific CUDA/Triton custom operators

📋 Requirements

• Experience in machine learning or high performance graphics
• Solid practical understanding of PyTorch or JAX
• Strong ability to write maintainable code in Python and/or C++
• Performance and efficiency oriented mindset

✨ Nice to Have

• Hands-on experience with torch.compile and FSDP2 APIs
• Experience building Triton kernels
• Experience with large-scale distributed training and parallelization techniques

🎁 Benefits & Perks

• 🏖️ Remote flexibility with on-site options
• 🌍 Global team of AI pioneers and graphics veterans
• 💡 Cutting-edge work on 3D generative AI frontier

📨 Hiring Process

Estimated timeline: 2-4 weeks

1. 1Recruiter call· 30 min
2. 2Online assessment· 1 hr
3. 3Technical coding interview· 1.5 hr
4. 4Offer· 1 week