Neural Network Performance Engineer

📋 Description

• Analyze performance bottlenecks of a model and propose improvements.
• Run the model on new hardware (e.g., NVIDIA Thor) efficiently.
• Implement custom kernels to reduce memory throughput where it matters.
• Quantize a model with minimal loss of quality.
• Suggest and implement model-architecture changes for better performance.

🎯 Requirements

• 3+ years building deep-learning systems with shipped models or publications.
• 1+ years in neural network inference performance (bottlenecks, kernels, quantization, compilers).
• Excellent understanding of GPU architecture and why some models run faster.
• Strong Python + PyTorch/JAX; able to profile, debug numerics, and write maintainable code.
• Document experiments clearly and communicate trade-offs crisply.
• Nice to have: robotics/autonomous driving and VLM/VLA familiarity.

🎁 Benefits

• Competitive equity: stock options with meaningful upside as we scale.
• 30+ paid days off, including 23 days of annual leave and UK bank holidays.
• Private healthcare, including virtual and in-person care.
• Pension scheme with 8% total contribution (5% employee, 3% employer).
• Free daily breakfast, catered lunch, and snacks in-office.
• Work with world-class engineers and leadership on AI humanoid robotics.