Staff / Principal Machine Learning Engineer (AI)

TL;DR

Staff/Principal Machine Learning Engineer (AI): Building and optimizing real-time multimodal model serving and orchestration platforms with an accent on inference optimization and high-performance systems. Focus on reducing latency, implementing advanced acceleration techniques, and scaling distributed GPU inference across multi-node clusters.

Location: Remote within Switzerland. Candidates must already have the legal right to work in Switzerland (no visa sponsorship available).

Company

An AI research lab developing best-in-class real-time multimodal models and a high-performance orchestration platform.

What you will do

• Optimize model serving using modern frameworks such as vLLM or TRT-LLM.
• Implement acceleration techniques including quantization, distillation, caching strategies, and speculative decoding.
• Develop high-performance systems using C++, CUDA, Rust, or highly optimized Python to maximize NVIDIA GPU utility.
• Scale distributed inference using Kubernetes and Ray, managing multi-GPU/multi-node setups for thousands of concurrent connections.
• Take full-cycle ownership of models from research to containerization and stable production serving.

Requirements

• Deep understanding of modern serving frameworks and inference optimization techniques.
• Proficiency in C++, CUDA, Rust, or optimized Python with strong profiling skills.
• Experience with Kubernetes, Ray, and custom load balancing for distributed systems.
• Legal right to work in Switzerland (mandatory).
• Professional English fluency (written and spoken) for collaboration with US-based teams.

Nice to have

• PhD in CS, Physics, Math, or equivalent practical experience building ML systems.
• Open-source contributions to major inference engines.
• Non-trivial systems programming projects or deep-dive technical write-ups.

Culture & Benefits

• Flat organizational structure with fast iterations and minimal process theater.
• High-impact environment where stability and shipping are prioritized over theoretical optimization.
• Culture of ownership where engineers are encouraged to question architecture and design benchmarks.
• Potential for future full U.S. visa and relocation support to the San Francisco Bay Area, subject to business needs.