Senior ML Engineer — Scale LLM Inference on GPU Cloud

United States Digital Space LLC; Amsterdam; Hybrid; EUR 100,000 - 180,000; Full time

Benefits offered by this job

Competitive compensation

Career growth and learning

Flexibility and ownership

Collaborative and innovative culture

Impactful AI projects

International environment and talented

Job summary

United States Digital Space LLC in Amsterdam is building a high-performance AI cloud platform that scales GPUs for training, inference, and deployment. Token Factory is part of Cloud, a major GPU cloud project, delivering throughput, latency, and cost improvements across thousands of GPUs.

You will join a team of engineers shaping large-scale inference engines, low-precision training, and open-source contributions, with opportunities for impact in a fast-moving startup-like environment.

Qualifications

• A profound understanding of theoretical foundations of machine learning and transformer architecture.
• Experience profiling GPU workloads using Nsight, PyTorch profiler, or similar tools
• Understanding of GPU memory hierarchy and compute/memory tradeoffs
• Familiarity with important ideas in LLM space, such as MHA, RoPE, KV-cache, Flash Attention, and quantisation
• Understanding of performance aspects of large neural network training (sharding strategies, custom kernels, hardware features etc.)
• Strong software engineering skills (we mostly use Python)
• Deep experience with modern deep learning frameworks
• Proficiency in contemporary software engineering approaches, including CI/CD, version control and unit testing
• Strong communication and leadership abilities

Responsibilities

• Inference Optimization: Identifying bottlenecks to drive production speedups and maximize performance.
• Inference engines support: Implement novel speculative decoding architectures and contribute to open-source inference engines.
• Low Precision Training & Inference: Design and productionise low-precision training/inference pipelines with measurable gains.

Skills

ML theory

Transformer architecture

Gpu profiling

GPU memory hierarchy

LLM concepts

Python

CI/CD

Leadership

Tools

Nsight

PyTorch profiler

CUDA

Job description

United States Digital Space LLC in Amsterdam is building a high-performance AI cloud platform that scales GPUs for training, inference, and deployment. Token Factory is part of Cloud, a major GPU cloud project, delivering throughput, latency, and cost improvements across thousands of GPUs.

You will join a team of engineers shaping large-scale inference engines, low-precision training, and open-source contributions, with opportunities for impact in a fast-moving startup-like environment.