🛠 Tech Stack

💼 About This Role

You'll lead inference optimization and model serving for a top AI research lab, building realtime multimodal systems. Your work directly powers products used by leading companies. You'll own full-cycle delivery from research to production.

🎯 What You'll Do

• Optimize inference serving frameworks like vLLM or TRT-LLM
• Profile and accelerate model performance on NVIDIA GPUs
• Design distributed systems for multi-GPU/multi-node inference
• Containerize and deploy models to production reliably

📋 Requirements

• Deep understanding of serving frameworks (vLLM, TRT-LLM)
• Hands-on experience with quantization, distillation, caching
• Proficiency in C++, CUDA, Rust, or optimized Python
• Experience with Kubernetes, Ray, and distributed scaling

✨ Nice to Have

• Non-trivial open-source contributions to inference engines
• PhD in CS, Physics, Math or equivalent practical experience
• Public technical write-ups or deep-dive systems projects

🎁 Benefits & Perks

• 💰 Competitive base salary $270k-$500k+ bonus+equity
• 🏢 Relocation assistance to Mountain View office
• 📈 Equity in a top AI startup backed by major VCs
• 🏥 Benefits package (not detailed but implied)