Robotics AI Inference Engineer

Thehumanoid; Greater London; On-site; GBP 65,000 - 85,000; Full time

Benefits offered by this job

23 days annual leave

Fully funded private healthcare

Pension scheme with 8% contribution

Free daily breakfast

Catered lunch and snacks

Job summary

Thehumanoid is seeking a talented Neural Network Performance Engineer to improve the performance of neural network-based control policies in London. This role involves enhancing model efficiency, analyzing performance bottlenecks, and implementing optimizations for neural networks.

The ideal candidate has significant experience in deep-learning systems, custom kernels, and GPU architecture, while contributing to innovative robot technology. Coming with competitive benefits including private healthcare and equity ownership, this position promises impactful contributions to AI research and robotics.

Qualifications

• 3+ years building deep-learning systems with shipped models.
• 1+ years experience in neural network inference performance.
• Strong understanding of GPU architecture.

Responsibilities

• Analyze performance bottlenecks of a model architecture.
• Make the model run efficiently on new hardware.
• Implement custom kernels to reduce memory throughput.

Skills

Deep-learning systems

Neural network performance analysis

GPU architecture understanding

Python

PyTorch

JAX

Job description

Thehumanoid is seeking a talented Neural Network Performance Engineer to improve the performance of neural network-based control policies in London. This role involves enhancing model efficiency, analyzing performance bottlenecks, and implementing optimizations for neural networks.

The ideal candidate has significant experience in deep-learning systems, custom kernels, and GPU architecture, while contributing to innovative robot technology. Coming with competitive benefits including private healthcare and equity ownership, this position promises impactful contributions to AI research and robotics.